1741_C16.qxd
2/11/2005
9:57 AM
Page 527
16
Polylactic Acid Technology
David E. Henton, Patrick Gruber, Jim Lunt, and Jed Randall
CONTENTS
16.1 Introduction ..............................................................................................528
16.2 Lactic Acid ................................................................................................531
16.3 Life Cycle Analysis ..................................................................................532
16.4 Polymerization of Lactide ......................................................................536
16.5 PLA Physical Properties ..........................................................................538
16.5.1 Linear Optical Copolymer Structures and Blends ..................538
16.5.1.1 Density ..........................................................................540
16.5.1.2 Glass Transition and the Amorphous Phase ............540
16.5.2 Rheology ........................................................................................542
16.5.2.1 Melt Rheology of Linear PLA ....................................542
16.5.2.2 Melt Stability ................................................................544
16.5.2.3 Solution Properties ......................................................544
16.5.2.4 Branching ......................................................................545
16.5.2.5 Solid-State Viscoelastic Properties ............................547
16.5.3 Crystallinity ..................................................................................547
16.5.3.1 Morphology ..................................................................547
16.5.3.2 Degree of Crystallinity ................................................550
16.5.3.3 Crystallization Kinetics ..............................................552
16.5.3.4 Stereocomplex ..............................................................553
16.5.3.5 Stress-Induced Crystallization ..................................554
16.5.4 Degradation and Hydrolysis ......................................................555
16.5.4.1 Hydrolysis ....................................................................555
16.5.4.2 Hydrolysis of Solid Samples Suspended in
Aqueous Media ............................................................558
16.5.4.3 Solution Hydrolysis ....................................................560
16.5.4.4 Hydrolysis of Samples Exposed to Humidity ........561
16.5.5 Enzymatic Degradation ..............................................................563
16.6 Applications and Performance ..............................................................563
16.7 Summary ....................................................................................................568
527
1741_C16.qxd
2/11/2005
9:57 AM
528
Page 528
Natural Fibers, Biopolymers, and Biocomposites
Acknowledgments ............................................................................................569
Reference ..............................................................................................................569
ABSTRACT Poly(lactic acid) (PLA) is the first commodity polymer produced from annually renewable resources. Some of the environmental benefits of PLA, such as low energy to produce and reduced green house gas
production, are presented and opportunities for the future are given. This
chapter on PLA reviews the chemistry and material properties of the polymer and its copolymers. The current technology for the production of lactic
acid, lactide, and polymer are presented. Applications for fabricated articles
such as films and fibers are also discussed.
16.1
Introduction
Polylactic acid (PLA) is a rigid thermoplastic polymer that can be semicrystalline or totally amorphous, depending on the stereopurity of the polymer backbone. L()-lactic acid (2-hydroxy propionic acid) is the natural and
most common form of the acid, but D()-lactic acid can also be produced by
microorganisms or through racemization and this “impurity” acts much like
comonomers in other polymers such as polyethylene terephthalate (PET) or
polyethylene (PE). In PET, diethylene glycol or isophthalic acid is copolymerized into the backbone at low levels (1–10%) to control the rate of crystallization. In the same way, D-lactic acid units are incorporated into L-PLA
to optimize the crystallization kinetics for specific fabrication processes and
applications.
PLA is a unique polymer that in many ways behaves like PET, but also
performs a lot like polypropylene (PP), a polyolefin. Ultimately it may be the
polymer with the broadest range of applications because of its ability to be
stress crystallized, thermally crystallized, impact modified, filled, copolymerized, and processed in most polymer processing equipment. It can be
formed into transparent films, fibers, or injection molded into blowmoldable preforms for bottles, like PET. PLA also has excellent organoleptic
characteristics and is excellent for food contact and related packaging applications.
In spite of this unique combination of characteristics, the commercial viability has historically been limited by high production costs (greater than
$2/lb). Until now PLA has enjoyed little success in replacing petroleumbased plastics in commodity applications, with most initial uses limited to
biomedical applications such as sutures.1
PLA is not new to the world of polymers. Carothers2 investigated the production of PLA from the cyclic dimer (lactide) of lactic acid as early as 1932.
Even before that, low molecular weight dimers and oligomers were detected
1741_C16.qxd
2/11/2005
9:57 AM
Page 529
Polylactic Acid Technology
529
when water was removed from an aqueous solution of lactic acid.3 The
announcement of the formation of a new company, Cargill Dow LLC, in 1997
brought two large companies together to focus on the production and marketing of PLA with the intention of significantly reducing the cost of production and making PLA a large-volume plastic.4
In today’s world of green chemistry and concern for the environment,
PLA has additional drivers that make it unique in the marketplace. The starting material for the final polymer, lactic acid, is made by a fermentation
process using 100% annually renewable resources. The polymer will also
rapidly degrade in the environment and the by-products are of very low toxicity, eventually being converted to carbon dioxide and water.
PLA can be prepared by both direct condensation of lactic acid and by
the ring-opening polymerization of the cyclic lactide dimer, as shown in
Figure 16.1. Because the direct condensation route is an equilibrium reaction,
difficulties of removing trace amounts of water in the late stages of polymerization generally limit the ultimate molecular weight achievable by this
approach. Most work has focused on the ring-opening polymerization of lactide, although other approaches, such as azeotropic distillation to drive the
removal of water in the direct esterification process, have been evaluated.5
Cargill Dow LLC has developed a patented, low-cost continuous process
for the production of lactic acid-based polymers.6 The process combines the
substantial environmental and economic benefits of synthesizing both lactide
and PLA in the melt rather than in solution and, for the first time, provides a
commercially viable biodegradable commodity polymer made from renewable resources. The process starts with lactic acid produced by fermentation of
dextrose, followed by a continuous condensation reaction of aqueous
lactic acid to produce low molecular weight PLA prepolymer (Figure 16.2).
Next, the low molecular weight oligomers are converted into a mixture of
lactide stereoisomers using a catalyst to enhance the rate and selectivity of the
H2O
O
O
O
HO
OH
H
H
CH3
L-PLA
O
H
L-Lactic acid
O
H2O
CH3
CH3
O
H
O
L-Lactide
FIGURE 16.1
Polymerization routes to polylactic acid.
n
CH3
1741_C16.qxd
2/11/2005
9:57 AM
Page 530
530
Natural Fibers, Biopolymers, and Biocomposites
O
OH
HO
O
HO
O
Lactic acid
OH
O
n
O
O
High molecular weight PLA
Condensation
−H2O
Ring-opening
polymerization
O
O
O
HO
O
n
O
O
OH
O
Depolymerization
O
O
Prepolymer
Mn ~ 5000
FIGURE 16.2
Schematic of PLA production via prepolymer and lactide.
Dextrose
Corn
Unconverted
monomer
PLA
polymer
Lactide
formation
Prepolymer
Meso
lactide
Polymerization
Low D
lactide
Distillation
Lactic
acid
Distillation
Fermentation
FIGURE 16.3
Nonsolvent process to prepare polylactic acid.
intramolecular cyclization reaction. The molten lactide mixture is then purified by vacuum distillation. Finally, PLA high polymer is produced using an
organo tin-catalyzed, ring-opening lactide polymerization in the melt, completely eliminating the use of costly and environmentally unfriendly solvents.
After the polymerization is complete, any remaining monomer is removed
under vacuum and recycled to the beginning of the process (Figure 16.3).
1741_C16.qxd
2/11/2005
9:57 AM
Polylactic Acid Technology
Page 531
531
This process is currently in operation in a recently constructed 300 million lb/yr
commercial-scale PLA plant in Blair, Nebraska, USA. Cargill Dow LLC has
also announced the construction of an additional plant in Europe in the near
future.7
PLA has been extensively studied by many researchers and reviewed in
detail in several recent publications.8 This chapter will not review all of the
growing volume of information on PLA, but will focus on the key elements
of technology that will change this polymer from a specialty material to a
large-volume commodity plastic.
16.2
Lactic Acid
Lactic acid (2-hydroxypropionic acid) is the basic building block for PLA. It
is a highly water-soluble, three-carbon chiral acid that is naturally occurring
and is most commonly found in the L() form. It is used as an acidulant in
foods, as a building block for biodegradable polymers (PLA), and is converted to esters and used as a green solvent for metal cleaning, paints, and
coatings. PURAC has been the world’s largest producer of lactic acid, but the
recent start-up of the Cargill Dow lactic acid plant, with a capacity of
400,000,000 lb, exceeds that of all producers combined. There are different
grades of lactic acid, each varying in purity depending on their end use
application. The lowest to highest purity is technical grade, edible grade,
pharmaceutical grade, and analytical grade. The lower grades contain sulfates, metals, amino acids, and various carbohydrates. With the need for
polymerization-grade lactic acid it has become clear that even the highest
purity grade, analytical grade, is not pure enough. The slight traces of carbohydrates and amino acids remaining in analytical grade cause color in the
process and even parts per million traces of cations such as Na lead to
racemization of the acid, resulting in the less desired D() lactic acid.
A good summary of lactic acid chemistry is found in the book by Holten,9
Lactic Acid, Properties and Chemistry of Lactic Acid and Derivatives. An older,
but still informative review of lactic acid production via fermentation is
found in a chapter on lactic acid by H. Schopmeyer.10 The major issues of fermentation, purification, and a history of lactic acid growth in the United
States are reviewed. It also contains a good literature review up to that time
on lactic acid.
A more recent review11 (1976) of lactic acid processes by Cox and Macbean
looked closely at the details of distillation, liquid extraction, esterification,
salt processes, electrodialysis, thermal methods, and ion exchange
approaches to providing lactic acid from fermentation of carbohydrates.
There have been a large number of proposed routes and patents issued
claiming to have improved processes to isolate lactic acid from fermentation
broth, but most are combinations or improvements on these basic concepts.
1741_C16.qxd
2/11/2005
9:57 AM
Page 532
532
Natural Fibers, Biopolymers, and Biocomposites
The main stages of lactic acid production consist of (1) fermentation, (2) cell
mass and protein removal, (3) recovery and purification of lactic acid,
(4) concentration of lactic acid, and (5) color removal.
Research and improvements in lactic acid production have been centered
around each of these steps, with continued improvement in the cost
and efficiency of each. Better microorganisms for the fermentation that are more
robust to temperature, pH, and the organic products they produce have evolved
in recent years. Engineering improvements have increased the size and efficiency of the plants, while improved technology for water removal (multiple
effect evaporators and membranes) has lowered the cost for production. By far
the greatest developments and diversity of approaches have been in the area of
separation and purification. Some of the separation and purification processes
and their proposed key features are listed in Table 16.1.
Most processes to produce lactic acid rely on bacterial fermentation of dextrose in aqueous slurry with a continual addition of a base, such as calcium
hydroxide, to maintain a neutral pH. Bacteria are generally quite sensitive to
pH and are only productive in a narrow pH range. This sensitivity to pH is
demonstrated in Table 16.2 for the production of acetic acid by Clostridium
thermoaceticum.12 Generally, deviation of pH by more than a couple of pH
units totally inhibits bacterial fermentation. For the production of neutral
species such as ethanol, this is not an issue. However, the production of
organic acids such as acetic and lactic acid shifts the pH as the acids are produced. To maintain the optimum pH, bases are continuously added.
At a pH of about 7 (neutral pH), essentially all of the lactic acid is ionized.
Isolation of the lactic acid and removal of water requires that a strong mineral acid, such as sulfuric acid, be added to protonate the lactate salt, while
at the same time producing a mole of salt for every two moles of acid produced. On a weight basis, about a pound of salt per pound of acid is produced. This significant amount of waste salt contributes to the cost of the
lactic acid and is undesirable from a sustainability point of view.
The process of the near future will contain an organism that is tolerant of
temperature and acidic conditions and be combined with lactic acid removal
technology that does not result in the generation of waste salts. This process
will be cost- advantaged and beneficial to the environment.
16.3
Life Cycle Analysis
In recent years, products, processes, and technology have been judged for their
overall sustainability. Those that use large quantities of energy, deplete natural
resources, pollute the environment, or emit greenhouse gases are viewed as less
sustainable. Green chemistry is the use of chemistry for pollution prevention or
the production of materials that fulfill the definition of sustainability. More
specifically, green chemistry is the design of chemical products and processes
The lactate salt is acidified by a strong
acid ion exchange resin.
Lactic acid is removed from the aqueous solution
by complexing with an amino containing resin.
Lactic acid is separated from less volatile
components by vacuum steam distillation.
The fermentation or purification process is run
at a concentration that exceeds the solubility
of the lactate salt (e.g., CaSO4), which is isolated
and later acidified.
Lactate esters are prepared and the
volatile esters are distilled.
Ion exchange (acidification)
Ion exchange (purification)
Distillation
Insoluble salt processes
1. Distillation and separation of esters
gives high-quality product.
2. Requires reconversion to acid.
1. Does not require acidification of fermentation.
2. Energy cost and capital.
1. Higher productivity because one can
maintain low acid level in fermenter.
2. Fouling of the membrane.
3. Requires acidic pH stable organism.
1. Suitable for continuous process and provides efficient
removal from many non-acidic impurities.
2. High cost of capital.
3. Solvent loss costs.
1. Eliminates the need to add a strong acid to the
fermentation.
2. Cost of resin and issues of resin regeneration.
1. This is the solid equivalent of t-amine extraction
technology without the solvent loss issues.
2. Regeneration of the resin.
3. Cost and availability of the resin.
1. Lactic acid can be steam distilled.
2. Significant purification must be done
prior to distillation.
3. Depending on conditions, some degradation
and oligomerization can occur.
1. Simple process that utilizes low-cost capital.
2. The crystallization of CaSO4 occludes impurities and
results in relatively impure acid.
Advantage/Disadvantage
9:57 AM
Esterification
Lactic acid is continuously removed from the fermentation
or acidified broth by preferential partitioning into a solvent.
Liquid extraction
Can be used to continuously remove lactic acid (lactate ions)
through a membrane driven by electrical potential.
Lactic acid is continuously removed
through a membrane.
Feature
2/11/2005
Reverse osmosis
Electrodialysis
Lactic Acid Purification Unit Operations
TABLE 16.1
1741_C16.qxd
Page 533
Polylactic Acid Technology
533
1741_C16.qxd
2/11/2005
9:57 AM
Page 534
534
Natural Fibers, Biopolymers, and Biocomposites
TABLE 16.2
Fermentation Performance of Clostridium
Thermoaceticum vs. pH in Acetic Acid Production
pH
Productivity (g/hr)
Acetate –g/l (as acid)
Yield (%)
6.8
0.8
45
85–90
5.5
0.6
16
77
that reduce or eliminate the use and generation of hazardous substances. In a
production supply chain, green products and processes integrate engineering,
chemistry, and environmental issues. The supply chain evaluation incorporates
the energy, wastes, and emissions for operations during the production and
transportation of all raw materials throughout the entire process, from production of all raw materials to disposal of the final articles.
An excellent method to quantify the sustainability of a product or process
is to do a life cycle analysis (LCA). LCA looks at all factors from raw materials, transportation, product manufacture, end use, to disposal. The energy,
air and water pollution, as well as the hazards of the chemicals are assessed
and judged against alternate products or solutions. Each step from cradle to
grave is quantified and the total impact on the environment in each of the
areas is thus known.
Cargill Dow’s process embodies the well-known Principles of Green
Chemistry by (1) preventing pollution at the source through the use of a natural fermentation process to produce lactic acid, (2) substituting annually
renewable materials for petroleum-based feedstock, (3) eliminating the use
of solvents or other hazardous materials, (4) completely recycling product
and by-product streams, and (5) efficiently using catalysts to reduce energy
consumption and improve yield. In addition, NatureWorksTM PLA products
can be either recycled or composted after use.
PLA, as produced by Cargill Dow LLC, uses from 20 to 50% less fossil
fuel than conventional plastic resins (Figure 16.4), meaning that up to 2 to 5
times more PLA can be produced with a given amount of fossil fuel than
petrochemical-derived plastics. The ability to produce the polymer on a
favorable cost/performance basis is critical to realizing these human health
and environmental benefits. This has been achieved through efficient use of
raw materials, reduced energy utilization, and elimination of solvents.
PLA is also a low-impact, greenhouse gas polymer because the CO2 generated during PLA biodegradation is balanced by an equal amount taken
from the atmosphere during the growth of plant feedstocks. In contrast,
petrochemical-based polymers contribute to volatile organic compound
(VOC) emissions and CO2 generation when incinerated. LCA calculations
indicate that PLA has a greenhouse gas emission rate of about 1600 kg
CO2/metric ton, while polypropylene, polystyrene, PET, and nylon have
greenhouse gas values of 1850, 2740, 4140, and 7150 kg CO2/metric ton,
respectively.13 Figure 16.5 shows the relative comparison of greenhouse gas
1741_C16.qxd
2/11/2005
9:57 AM
Page 535
Polylactic Acid Technology
160
535
142
140
MJ/kg
120
91
92
100
87
77
80
76
57
60
34
40
5
20
te
rm
ar
5
ng
er
ye
Lo
A
PL
A
or
ph
.
ye
ar
1
PE
C
PL
PP
PP
S
T
am
ph
el
N
lo
G
an
e TM
IP
S
H
yl
on
66
0
FIGURE 16.4
MJ/kg of energy required to produce various polymers (From Gruber, P., Keynote address,
Massachusetts Green Chemistry Symposium, Amherst, MA, 2001. With permission.)
14000
13030
12000
10000
8000
6870
6000
4143
4000
2737
2561
1852
2000
1820
−81
0
Nylon 66
HIPS
CellophaneTM GPPS
PP
PETAM
−1500
PLA year 1 PLA year 5 PLA long
term
−2000
FIGURE 16.5
Greenhouse gas emissions (kg/ton) for production of various polymers. Long-term emissions
for PLA are based on utilization of biomass for the production of lactic acid.
1741_C16.qxd
2/11/2005
9:57 AM
Page 536
536
Natural Fibers, Biopolymers, and Biocomposites
contribution to the environment for common plastic products. Longer term,
as PLA is produced from field wastes or other biomass, PLA can become a
CO2 sink and actually contribute to a net reduction in greenhouse gases.
Although problematic in the past, recycling can offer an environmentally
attractive disposal option for plastics. However, recycled plastics are typically
contaminated with other incompatible materials and thus have reduced
properties. Consequently, recycled plastics are often relegated to lowperformance/low-value applications. In contrast, PLA articles can be readily
hydrolyzed with water to form lactic acid, which is then purified and polymerized to remake prime polymer. PLA can also be naturally composted into
humus, carbon dioxide, and water, with complete degradation occurring within
only a few weeks under typical compost conditions. However, polymer hydrolysis and monomer recovery offer the preferred method for PLA recycling.
Polymerization of Lactide
16.4
Many catalyst systems have been evaluated for the polymerization of lactide
including complexes of aluminum, zinc, tin, and lanthanides. Even strong
bases such as metal alkoxides have been used with some success. Depending
on the catalyst system and reaction conditions, almost all conceivable mechanisms (cationic,14 anionic,15 coordination,16 etc.) have been proposed to explain
the kinetics, side reactions, and nature of the end groups observed in lactide
polymerization. Tin compounds, especially tin(II) bis-2-ethylhexanoic acid (tin
octoate), are preferred for the bulk polymerization of lactide due to their solubility in molten lactide, high catalytic activity, and low rate of racemization of
the polymer. Conversions of 90% and less than 1% racemization can be
obtained while providing polymer with high molecular weight.
The polymerization of lactide using tin octoate is generally thought to occur
via a coordination-insertion mechanism16 with ring opening of the lactide to
add two lactyl units (a single lactide unit) to the growing end of the polymer
chain shown schematically in Figure 16.6. The tin catalyst facilitates the
O
R-OH
O
O
O
O
O
O
O
O
O
Sn(Oct)2
O
H
O
Sn
R
(Oct)2
O
O
O
O
O O
Sn R
H
O
H
RO
Sn(Oct)2
O
O
O
H
Sn(Oct)2
RO
(Oct)2
FIGURE 16.6
Generalized coordination-insertion chain growth mechanism of lactide to PLA: R, growing
polymer chain.
1741_C16.qxd
2/11/2005
9:57 AM
Page 537
Polylactic Acid Technology
537
polymerization, but hydroxyl or other nucleophilic species are the actual initiators. There is generally several hundred parts per million of hydroxyl impurities in the lactide from water, lactic acid, and linear dimers and trimers.
High molecular weight polymer, good reaction rate, and low levels of
racemization are obtained with tin octoate-catalyzed polymerization of lactide. Typical conditions for polymerization are 180°–210°C, tin octoate concentrations of 100–1000 ppm, and 2–5 h to reach 95% conversion. The
polymerization is first order in both catalyst and lactide. Frequently
hydroxyl-containing initiators such as 1-octanol are used to both control
molecular weight and accelerate the reaction.
Copolymers of lactide with other cyclic monomers such as caprolactone17
can be prepared using similar reaction conditions (Figure 16.7). These
monomers can be used to prepare random copolymers or block polymers
because of the end growth polymerization mechanism. Cyclic carbonates,
epoxides, and morpoline diones have also been copolymerized with lactide.
In addition to the work done on catalysts and comonomers, a significant
amount of work has been done designing the optimum polymerization
process from a cost and versatility perspective. Most of this work has used
lactide dimer as the starting point. Batch polymerization as well as continuous processes can be used. Continuous processes (Figure 16.3) have a cost
and productivity advantage and thus are the focus of most work. Stirred
tank or pipe reactors have been evaluated alone as well as in combination.
Because of the low energy of the ring-opening polymerization and the
potential to obtain high rates of polymerization, melt extruders have been
extensively evaluated as reactors to produce PLA. A number of groups have
reported or patented specific aspects of the use of extruders or combinations
of several reactor concepts to produce PLA.18–31
DuPont19 recognized the importance of high rates of lactide polymerization and the economic viability of using extruders. They investigated the
issues surrounding hydroxyl initiators, catalysts, and acidic impurities.
DuPont’s patent19 (US 5,310,599) provides examples showing the dramatic
(negative) effect of acidic impurities, such as the linear dimer (DP2) on the
rate of lactide polymerization. The acid impurities coordinate with the tin
catalyst and render it ineffective as a ring-opening polymerization site. As an
example, 389,000 MW PLA can be made with a 5-min residence time (6000:1
O
O
O
O
O
O
Sn(Oct)2
O
Lactide
O
O
+
ε-Caprolactone
FIGURE 16.7
Copolymerization of lactide and caprolactone.
On
m
1741_C16.qxd
2/11/2005
538
9:57 AM
Page 538
Natural Fibers, Biopolymers, and Biocomposites
monomer: Sn octoate) at 180oC in the absence of acidic impurities, and 98%
conversion is obtained. The conversion in 5 min drops to 50% at an impurity:catalyst ratio of 2:1 and to 12% in 5 min at a 6:1 ratio.
The rate of polymerization of lactide can be accelerated by the use of
hydroxyl initiators such as butanol or by higher catalyst loading. However,
increased initiator levels will reduce the molecular weight while more catalyst will reduce the thermal stability and increase the color of the PLA.
16.5
PLA Physical Properties
Several excellent reviews have been written which include details on the
properties and characteristics of PLA.32–36 A discussion of the variety of
structures and their thermal and mechanical behavior follows.
The physical characteristics of high molecular weight PLA are to a great
extent dependent on its transition temperatures for common qualities such
as density, heat capacity, and mechanical and rheological properties. In the
solid state, PLA can be either amorphous or semicrystalline, depending on
the stereochemistry and thermal history. For amorphous PLAs, the glass
transition (Tg) determines the upper use temperature for most commercial
applications. For semicrystalline PLAs, both the Tg (~58°C) and melting
point (Tm), 130°–230°C (depending on structure) are important for determining the use temperatures across various applications. Both of these
transitions, Tg and Tm, are strongly affected by overall optical composition,
primary structure, thermal history, and molecular weight.
Above Tg amorphous PLAs transition from glassy to rubbery and will
behave as a viscous fluid upon further heating. Below Tg, PLA behaves as a
glass with the ability to creep until cooled to its β transition temperature of
approximately 45°C. Below this temperature PLA will only behave as a
brittle polymer.
16.5.1
Linear Optical Copolymer Structures and Blends
Having two optical arrangements for lactic acid (L-lactic acid, D-lactic acid),
and three optical arrangements for lactide (L-lactide, D-lactide, meso-lactide),
the variety of primary structures available for PLA is substantial. In addition, many other monomers have been copolymerized with lactic acid
and/or lactide, the subject of which will not be covered in this chapter.
Techniques for measuring PLA stereosequences by NMR have been published by Thakur et al.37
Containing the discussion to strictly PLA linear optical copolymers, there are
several notable primary structures which lead to a variety of highly ordered
crystalline structures. The simplest is that of the isotactic homopolymer
1741_C16.qxd
2/11/2005
9:57 AM
Polylactic Acid Technology
Page 539
539
poly(L-lactide) (PLLA) or its opposite enantiomer poly(D-lactide) (PDLA)
where all of the monomers in the chain are of the same optical composition. A
greater amount of research has been done on PLLA due to the fact that the commercial production of lactic acid has been predominantly L-lactic acid. The crystal structure of PLLA homopolymer exists in α and β forms. A more detailed
description of the crystal structures and unit cell parameters follows.
The most common commercial polymers of PLA are optical copolymers of
predominantly L-lactide, with small amounts of D- and meso-lactides, made
through bulk polymerization with tin octoate catalyst via ring-opening
polymerization (ROP). While these copolymers are generally described as
random, there is evidence of some degree of stereoselectivity. Selectivity of
tin octoate is discussed by Thakur et al.38 Condensation polymerization of
mostly L-lactic acid with small amounts of D-lactic acid, polymerized in solution, has also been used. The optical comonomers introduce “kinks” in
PLA’s natural helical conformation and “defects” in the crystal arrangement,
which results in depression of the melting point, reduction in the level of
attainable crystallinity, and reduction in the rate of crystallization. Optical
purity (OP) is a common nomenclature to describe polymers of this variety.
As OP decreases, crystallization eventually becomes impossible and the
polymer is amorphous (occurring about when OP 0.78). For the purposes
of this chapter, the designation a-PLA will be used for PLAs made with
insufficient optical purity to crystallize, and c-PLA will be used to designate
PLAs that are crystallizable. Stereoselective catalysis has been used to
develop optical copolymers having a highly tapered concentration of optical
purity (OP) along the chain.39
Random copolymers made from meso-lactide result in an atactic primary
structure referred to as poly(meso-lactide) and are amorphous. Random optical copolymers made from equimolar amounts of D-lactide and L-lactide are
commonly referred to as PDLLA or poly(rac-lactide). PDLLA is also essentially atactic, but the primary structure is segregated into optical doublets of
the lactyl group, and it is also amorphous.
Ovitt and Coates40 have made two novel stereoregular primary structures
using chiral catalysts. The first was syndiotactic PLA, made by ROP of mesolactide with a chiral aluminum alkoxide catalyst, which selectively opened
the L-lactyl side of the meso-lactide molecule and resulted in alternating stereocenters along the growing chain. This can be visualized as an optical headto-tail addition and results in an –LDLDLD- arrangement of lactyl
stereocenters along the chain. In this case, optical purity (OP) will refer to the
fraction of stereoregular syndiotactic sequences. The material was shown to
crystallize with a melting point of about 152°C with an OP of 0.96. The second structure, referred to as di-syndiotactic PLA or heterotactic PLA, was
made by ROP of meso-lactide with a racemic catalyst that would add mesolactide. However, the reacted lactide stereocenter would add in an alternating
fashion, in what can be visualized as an optical head-to-head addition resulting in sequences of –LLDDLLDD- along the chain. The heterotactic PLAs produced did not crystallize, and had a Tg of about 40°C. Ovitt also showed that
1741_C16.qxd
540
2/11/2005
9:57 AM
Page 540
Natural Fibers, Biopolymers, and Biocomposites
racemic catalysts can be used to polymerize rac-lactide into stereoblock
copolymers that have long sequences of poly(L-lactide) and poly(D-lactide)
blocks, referred to as isotactic stereoblock PLA.41 Similar work had been performed earlier by Yui et al.42 With sufficiently long isotactic run lengths, these
polymers are able to crystallize into the stereocomplex crystal arrangement.
Blends of PLA optical copolymers have also offered a variety of new properties. The most notable is that of the stereocomplex PLA, which is made by
solvent or melt blending PLLA and PDLA where the polymers undergo stereocomplexation or racemic crystallization. This phenomenon has been an area
of intense study since first described by Ikada et al.43 A more detailed description of stereocomplex will follow. A second blend receiving attention is the
blend of crystallizable PLA (c-PLA) with noncrystallizable PLA (a-PLA).
Key melting point and glass transition data for PLA structures and blends
are listed in Table 16.3.
16.5.1.1 Density
Volumetric measurements show that PLA and lactide isomers in the liquid
state have approximately the same density as a function of temperature,
with a thermal coefficient of expansion α1 7.4 104°C1. The melt density
can be approximated using Equation (16.1) by Witzke:32
ρ150°C
ρ (g/cm3) 1α1(T(°C)150)
(16.1)
where ρ150 1.1452.
The solid amorphous PLA density is ~1.25 g/cm3. The purely crystalline
PLLA density32 is estimated to be 1.37–1.49 g/cm3.
16.5.1.2 Glass Transition and the Amorphous Phase
The PLA glass transition (Tg) increases with Mn and OP. Orientation and
physical aging also affect the Tg. The infinite molecular weight Tg for PLLA,
poly(meso-lactide) and PDLLA were estimated to be 61°, 46°, and 53°C,
respectively, by Witzke,32 who developed an equation describing the Tg of
unoriented poly(L-lactide-co-meso-lactide), Equation (16.2)
180,000
Tg 45 16 wL-mer7 wmeso
Mn
(16.2)
where wL-mer and wmeso are the starting mole fractions of L-lactide and mesolactide, respectively.
Physical aging affects the glass transition of PLA. It can occur between
about 45°C and Tg.32 The free volume reduction, characteristic of physical
aging, is observed by DSC as an endothermic volume expansion immediately
after Tg and is often referred to as the enthalpy of relaxation. Crystallization
and orientation reduce the rate and extent of PLA aging. The rate of aging
increases with temperature up to the Tg.
Random level of meso or
D-lactide in L-lactide
or D-lactic acid in L-lactic acid
LLLLLLLL mixed with
DDDDDDDD
LLLLLLLLLLLDDDDDDDD
Random optical
copolymers
PLLA/PDLA
stereocomplex
PLLA/PDLA stereoblock
complexes
Syndiotactic poly(meso-)
PLA
40
Ovitt and Coates (1999)
46
Witzke (1997)
53
Witzke (1997)
No stereocontrol
No stereocontrol
Atactic PDLLA
40
Ovitt and Coates (1999)
65–72
Tsuji and Ikada (1999)
45–65
55–65
Tg(°C)
LLDDLLDDLLDDLLDD
Al-centered rac- chiral catalyst
279
Tsuji and Ikada (1996)
205 Tsuji and Ikada (1995)
211 Runt (2001)
215 Kalb and Pennings
(1980)
165–210
Runt (2001)
Tm° (°C)
Heterotactic
(disyndiotactic) poly(mesolactide)
Atactic poly(meso-lactide)
220–230
Ikada et al. (1987)
205 Yui et al. (1990)
179 Ovitt and Coates (2000)
152
Ovitt and Coates (1999)
130–170
170–190
Tm (°C)
9:57 AM
DLDLDLDLDLDL
Al-centered R- chiral catalyst
LLLLLLLLL or DDDDDDDDDD
Isotactic poly(L-lactide) or
poly(D-lactide)
Description
2/11/2005
PLA Structure
Melting Point and Glass Transition Data for PLA Structures and Blends
TABLE 16.3
1741_C16.qxd
Page 541
Polylactic Acid Technology
541
1741_C16.qxd
2/11/2005
9:57 AM
542
Page 542
Natural Fibers, Biopolymers, and Biocomposites
A significant reduction in the Tg is found for PLLA crystallized under partially constrained conditions, where the polymer is prevented from shrinking.44 It was proposed that the constrained crystallization process increases
the net free volume in the amorphous phase. Tg depression increased with
increasing crystallinity. In contrast, when samples are allowed to shrink or
crystallize quiescently, Tg increases with crystallinity. It was shown that the
Tg could vary by as much as 30°C between the two methods, while the melting point (Tm), heat of fusion (∆Hm), and overall degree of crystallinity (xc)
were not influenced. It is possible that the reduction in Tg under partially
strained conditions may be related to suppressing physical aging.
Blends of PLAs of different optical compositions have been studied for miscibility and compatibility. The most notable blend is that of the PLA stereocomplex, which occurs when PLLA is mixed with PDLA and the
homopolymers cocrystallize into a new arrangement with a markedly higher
melting point. The interesting features of the stereocomplex crystal will be discussed later. Tsuji and Ikada45 reported that the Tg of stereocomplex PLA is 5°C
higher than that of the nonblended components. This was attributed to the
strong interaction between the L-lactyl and D-lactyl unit sequences in the amorphous region of the blends, resulting in a more dense chain packing. A second
very interesting blend is to blend a crystallizable polymer (c-PLA) of high optical purity (OP) with a polymer of low OP (a-PLA). Urayama et al.46 investigated blends of PLLA with PDLLA by dynamic mechanical analysis (DMA).
Tg behavior supported compatibility of both polymers. The modulus remained
constant regardless of blend ratio. Modulus changes were consistent with
homopolymers, where the E of the blends dropped from 2–3 109 Pa to a rubbery plateau of 1–3 106 Pa across the Tg for amorphous specimens. The rubbery plateau increased to a different level depending on the crystallinity of the
blend. Blends of this type provide a route to modify flexibility of the polymer
above the glass transition by controlling the overall level of crystallinity (xc).
16.5.2
Rheology
Rheological properties are useful in evaluating thermoplastics for their performance during processing operations. The literature reports a wide variety of
processes and products using PLA, spanning from high orientation processes
such as fiber spinning and oriented films to lower orientation processes such as
thermoforming and foams. The time scales of these processes are very different. Considerations for material flow characteristics, orientation, and crystallization must be made to gauge and predict the resulting properties of the
products made from these processes. Intrinsic factors affecting the flow characteristics of PLA are molecular weight distribution, degree and type of branching, optical composition, optical block length distributions, and melt stability.
16.5.2.1 Melt Rheology of Linear PLA
PLA, like many thermoplastics, is a pseudoplastic, non-Newtonian fluid.
Above the melting point, PLA behaves as a classic flexible-chain polymer
1741_C16.qxd
2/11/2005
9:57 AM
Page 543
Polylactic Acid Technology
543
across all optical compositions.47 PLA’s zero shear viscosity and shear thinning behavior have been measured by a variety of groups. However, complicated by the array of optical copolymers studied, the difficulty associated
with measuring the absolute molecular weight of PLA, and the issues of
melt stability, the interpretation of PLA’s rheology has not always been consistent.
The most comprehensive studies of linear PLAs in the melt were performed by Janzen and Dorgan48–50 who showed that zero shear viscosity
(ηo) scales very close to the usual 3.4 power with Mw, and that PLA obeys
the Cox-Merz rule well into the shear thinning region when in the linear
viscoelastic range. The plateau modulus for PLLA is ~1 MPa, which corresponds to a critical molecular weight of about 9000 g/mol. Previous reports
by Cooper-White and MacKay51 measured the rheological properties of
PLLA with molecular weights ranging from 2,000 to 360,000 and reported a
dependence of zero shear viscosity (ηo) on chain length to the 4.0 power
(Mw4.0). However, only a very small number of samples were used to build
the 4.0 power relationship. Dorgan also showed that PLA exhibits strain
hardening during extension in the melt, a characteristic necessary for fiber
spinning.
The complex viscosity for PLA with 213,000 Mw (relative to polystyrene)
and 13.0% D was measured at many temperatures and shear rates. The
shifted data shown in Figure 16.8 depict rheological behavior across six
orders of magnitude in shear for 180°C.
1.0E + 09
1.0E + 08
1.0E + 07
1.0E + 06
1.0E + 05
1.0E + 04
G' (Pa)
1.0E + 03
G'' (Pa)
1.0E + 02
|eta*| (Pa*s)
1.0E + 01
fit
1.0E + 00
1.0E − 01
1.0E − 02
1.0E − 03
1.E − 02 1.E + 00
1.E + 02
1.E + 04
1.E + 06
1.E + 08
1.E + 10 1.E + 12
aT ω (rad/s)
FIGURE 16.8
Melt rheology of 213.000 Mw (relative to polystyrene) 13.0%D PLA shifted to 180°C.
*Measurements by Jay Janzen, Colorado School of Mines.
1741_C16.qxd
544
2/11/2005
9:57 AM
Page 544
Natural Fibers, Biopolymers, and Biocomposites
16.5.2.2 Melt Stability
Issues with PLA melt stability during processing and rheological testing
have been cited by Witzke,32 Dorgan et al.,48 Ramkumar and Bhattacharya,52
and Feng and Hanna.53 Reactions leading to poor melt stability include chain
scission due to hydrolysis, chain scission due to thermal instability, and lactide formation by the well-documented “back-biting” mechanism.
Hydrolysis can be reduced or essentially eliminated with conventional drying operations where water is removed prior to melting. Recommended drying conditions are available for commercial PLAs. When PLA is exposed to
a humid environment following drying, moisture is quickly absorbed into
the polymer up to its equilibrium level of swelling. Kinetics for melt hydrolysis depend on water concentration, acid or base catalysis, and polymerization catalyst concentration.
The proton in the CH group of the main chain of PLA is labile. It has been
suggested that the proximity of this labile proton to the ester group affects the
thermal sensitivity of the polymer.52 The proton may be abstracted resulting
in the breakdown of the molecular chain. In practical extrusion processes,
thermal stability becomes important at temperatures greater than 250°C.
Using PLA polymerized with tin octoate catalyst, Cicero et al.54 showed that
the melt stability of PLA can be improved by the addition of small amounts of
tris(nonylphenyl)phosphite (TNPP), likely due to balancing chain extension
with degradation reactions. Adding TNPP during rheological testing greatly
stabilized the polymer and lengthened the time available for testing.
The thermodynamic equilibrium between polylactide and lactide
monomer varies with temperature and is about 4.3% lactide at 210°C.32 Since
lactide imparts undesirable characteristics during processing, such as fuming, it is desirable to depress the lactide formation kinetics and stabilize the
polymer. Gruber et al.55 showed that lactide formation could be substantially
reduced under normal processing by complexing the catalyst with additives,
resulting in a melt-stable PLA polymer.
16.5.2.3 Solution Properties
The dilute-solution properties of PLA are critical for characterizing molecular weight, chain flexibility, and defining solvent quality. Factors reported to
affect PLA’s intrinsic viscosity ([η]) are molecular weight and distribution,
optical composition, and branching. A recent publication submission by
Janzen et al.47 systematically measured the dilute-solution properties of
PLAs of wide OP in three solvent systems. Mark-Houwink parameters were
determined by comparing [η] to a broad range of molecular weights determined by light scattering in a mixed solvent system first used on PLA by
Kang et al.56 Tables 16.4 and 16.5 are taken from Janzen’s publication submission. Surprisingly, no effect of optical composition was found to affect
[η], within the error of the experiments. Earlier work by Schindler and
Harper57 had shown a significant difference between the Mark-Houwink
parameters of PLLA and PDLLA.
1741_C16.qxd
2/11/2005
9:57 AM
Page 545
Polylactic Acid Technology
545
TABLE 16.4
Dilute-Solution Constants for PLAs at 30°C
Parameters
Schulz-Blaschke
concentration factor
Mark-Houwink fit
Stockmayer-Fixman fit
Characteristic ratios
Source:
kSB:
K, mL/g:
a:
Ke, mL/g:
104b:
C:
CHCl3
Solvent
THF
CH3CN &
CH2Cl2 Mix
0.302 0.005
0.289 0.005
0.373 0.016
0.0131 0.0048
0.777 0.031
0.112 0.017
5.82 0.69
6.74 0.67
0.0174 0.0046
0.736 0.023
0.101 0.014
4.76 0.65
6.29 0.57
0.0187 0.0075
0.697 0.034
0.096 0.011
2.30 0.36
6.08 0.46
Janzen, J. et al., submitted to Macromolecules, Feb. 11, 2003. With permission.
TABLE 16.5
Flexible Linear Polymer Melt Chain Dimension Comparisons
Values
Parameters
PLA (140°C)
PEO (140°C)
it-PP (190°C)
M1, relative molecular mass
per backbone bond
L2, Å2
ρ, g/cm3
GN0, MPa
Me, entanglement molecular mass
Me/M1, backbone bonds
Mc, critical molecular mass
Mc /M1, backbone bonds
r20/M, Å2
C
p, packing length, Å
dt, tube diameter, Å
dt/p (Ronca-Lin ratio)
24.02
14.68
21.04
Source:
2.05
1.152
1.0
3,959
165
9,211
383
0.574
6.7
2.51
47.7
19
2.11
1.034
1.8
1,973
134
5,330
363
0.805
5.6
1.99
39.9
20
2.37
0.759
0.43
6,797
323
13,635
648
0.694
6.2
3.15
68.7
21.8
PE (140°C)
14.027
2.42
0.792
2.6
1,046
75
3,164
226
1.25
7.3
1.68
36.2
21.6
Janzen, J. et al., submitted to Macromolecules, Feb. 11, 2003. With permission.
Calculated values for some key PLA parameters, including entanglement
molecular weight, critical molecular weight, and reptation parameters
(Reference 47) are shown in Table 16.5. The characteristic ratio (C) for PLA is
6.74 0.67. In sharp contrast to this value for C, Kang et al.56 characterized
PLA of varying OP by static and dynamic laser light scattering in conjunction
with Raman spectroscopy, and calculated a C of 12.4 for PLLA. Some debate
still remains as to the flexibility of PLAs of varying optical composition.
16.5.2.4 Branching
With the introduction of branching, the rheological properties of PLA can be
significantly modified. Since the linear polymer exhibits low melt strength,
1741_C16.qxd
2/11/2005
9:57 AM
Page 546
546
Natural Fibers, Biopolymers, and Biocomposites
for certain applications it is desirable to increase the melt strength by introducing long-chain branching. Several routes have been used to obtain longchain branching. Figure 16.9 shows the effect of branching on the complex
viscosity of PLA.
1. Multifunctional polymerization initiators have been used to make
star- and comb-shaped molecules. A number of studies have examined reacting lactides with branched polyols since first proposed
by Schindler et al.58 Dorgan et al.59 published rheological characterization of linear and star PLAs. Relaxation spectra show that the
transition zones for the linear and branched materials are nearly
indistinguishable, while the star polymers have greater contributions to the terminal regime and a much steeper dependence of ηo
as a function of Mw.
Zhao et al.60 made starburst structures by polymerizing lactide
onto dendrimer macroinitiators. From 16 to 21 polylactide arms
were attached to the surface of the initiator. Molecular weight is
controlled by the ratio of monomer to initiator.
2. Hydroxycyclic ester and/or carbonate polymerization initiators
are comprised of an initiation group (e.g., hydroxyl) and a ring
capable of copolymerizing with a growing PLA chain. Drumright
et al.61 reported polymerizing PLA with hydroxycarbonate initiators, resulting in long-chain branching and enhanced melt elasticity. Tasaka et al.62 made branched PLA with mevalonic lactone
initiator.
Complex viscosity (poise)
105
104
Peroxide-modified PLA
Linear PLA
1000
0.01
0.1
1
Frequency (rad/s)
10
FIGURE 16.9
Complex viscosity of branched PLA made by peroxide treatment.
100
1741_C16.qxd
2/11/2005
9:57 AM
Polylactic Acid Technology
Page 547
547
3. Multicyclic ester, multicyclic carbonate and/or multicyclic epoxy
comonomers are comprised of multiple rings (most commonly
two) capable of copolymerizing with a growing PLA chain.
Drumright et al.63 also reported on polymerizing PLA with bicyclic
diesters and/or dicarbonates to control rheological properties.
The bicyclic diesters and dicarbonates copolymerize easily with
lactide to form copolymers that can have tailored levels of branching. The copolymers showed excellent rheological properties,
including increased melt tensions and improved shear thinning,
compared to the analogous linear polymers. A typical polymer was
manufactured by polymerization of L-lactide at 130°–185°C with
0.1% α,α-2,5-dioxabicyclo[2.2.2]octane-3,6-dione. Since multicyclic
lactones do not contain an initiator, the chemistry can theoretically
be used to cross-link PLA to infinite molecular weight. A patent by
Gruber et al.64 describes rheological modification of PLA by copolymerizing with multifunctional expoxidized oil.
4. Cross-linking via free radical has been accomplished in the melt by
treating PLA with peroxides.65 Small amounts of peroxide result in
long-chain branches. The polymer remains melt processable and is
stable. Large improvements in melt elasticity have been measured
using die swell, melt tension, and parallel plate rheometry. Blends
of linear and branched PLAs showed good compatibility and predictable melt rheology by the mixing rule.66
16.5.2.5 Solid-State Viscoelastic Properties
Solid-state rheological properties of amorphous PLA typically show a storage modulus (E’) of about 2–3 109 Pa below the glass transition temperature. As the sample is heated, a sharp drop in modulus occurs between 55°
and 80°C, and E’ decreases to ~1–3 106 Pa as the glass transition (Tg) is
exceeded.67 Semicrystalline samples show less of a drop in E’ across the Tg,
and also a higher Tg with increasing crystallinity. Peak height of the tan δ
curve will decrease, but the width of the tan δ will increase as the percentage
of crystallinity increases. Highly crystalline samples will typically show a
tan δ peak at 70°–80°C. Optical composition and crystallization conditions
will strongly influence the onset of melting, and the corresponding drop in
E’. The β transition of PLA is very small and is scarcely detectable by DMA.
16.5.3
Crystallinity
16.5.3.1 Morphology
The morphology of PLA crystals is influenced by composition and thermal
history. Abe et al.68 showed that PLLA crystals grown at different temperatures go through three regime changes associated with nucleation and crystal growth rates. A morphology change was observed between regimes I and
II. For PLLA crystallized isothermally from the melt, spherulitic morphology
1741_C16.qxd
548
2/11/2005
9:57 AM
Page 548
Natural Fibers, Biopolymers, and Biocomposites
is observed at crystallization temperatures below 145°C (regime II). During
isothermal growth, the radius of the spherulites increases linearly with
time, and the slope of the curve is the lineal crystal growth rate (LCR). At
higher temperatures, a smaller number of spherulites are formed due to a
decrease in the nucleation density. This leads to larger spherulites with
increasing crystallization temperature. Crystallizing PLLA at temperatures
greater than 150°C (regime I) results in hexagonal lamellar stacking crystal
morphology.
Spherulites have been shown to contain stacked lamellar morphology,
with the long axis of the lamellar crystal running parallel to the spherulite
radius. Within the crystal, PLA exists as a polymeric helix, with an
orthorhombic unit cell with dimensions a 1.070 nm and b 0.614 nm.
Amorphous polymer resides between the lamellae. The lamellar thickness of
PLA varies with the temperature of crystallization (Tc), crystallization time
(tc), optical purity (OP), and molecular weight. PLLA grown at 120°, 140°,
and 160°C had lamellar periodicities measured at 14–16, 16–18, and
18–22 nm, respectively, measured by AFM.68
For L-rich polymers with small amounts of D- and meso-lactides, there is an
observed decrease in the equilibrium melting point with decreasing OP. At
least two possible explanations for this have been considered. In one extreme
case it is assumed that the optically impure units are incorporated into the
crystal structure as defects. This results in a decreasing melting point (Tm)
due to the decrease in the enthalpy of fusion. In the other extreme case, it is
assumed that the optically impure units are rejected from the crystal structure, causing a reduction in Tm due to entropy effects. Baratian et al.69 studied the degree of crystallinity, spherulitic growth rate, lamellar thickness,
and melting temperature of a series of optical copolymers of this variety,
comparing the effects of D-lactide and meso-lactide optical impurities. Smallangle x-ray scattering (SAXS) results showed that lamellar thickness
decreases with decreasing OP at a given degree of supercooling. It was proposed that a critical sequence length that is free of stereo defects is necessary
for crystallization and that the stereo defects are excluded from the crystalline regions. In a different study by Zell et al.,70 it was shown that, at least
under long crystallization times, a certain fraction of optical impurity (in this
case, a small percentage of labeled L-lactide copolymerized into a polymer of
predominantly D-lactide) is incorporated into the crystal lamellae. Zell
et al.70 went on to develop a model for determining the fraction of crystallizable PLA based upon a critical chain length necessary for crystallization
that allows stereo defects to be included in the crystalline region if they are
surrounded by a sufficient isotactic run length.
The spherulite structure for L-lactide-rich PLA was studied by Pyda et al.71
Figure 16.10 shows the crystal imaging using AFM and polarized light
microscopy.
A graph showing the observed melting point by DSC after isothermal
treatments at crystallization temperature Tc is shown in Figure 16.11 for a
series of poly(L-lactide-co-D-lactide)s that are rich in L-lactide content.69
Polymers were made by ROP using tin octoate catalyst.
1741_C16.qxd
2/11/2005
9:57 AM
Polylactic Acid Technology
Page 549
549
FIGURE 16.10
Morphology of PLA by optical microscopy and AFM. (a) Molecular formula of PLA. (b) Picture
of the central region of a spherulite by optical microscopy. (c) Copy of ab AFM picture of a
section of the spherulite shown in (b). (d) Schematic drawing of the lamellae within a
spherulite. (From Pyda et al., Proceedings of the 30th NATAS Annual Conference on Thermal
Analysis and Applications, 463, 2002. With permission.)
Using the Hoffman-Weeks72 method for determining equilibrium melting temperature (Tm0 ) of a polymer [linear extrapolation of experimentally observed melting points (Tm) vs. crystallization temperature (Tc)
1741_C16.qxd
2/11/2005
9:57 AM
Page 550
550
Natural Fibers, Biopolymers, and Biocomposites
200
190
180
PLLA
PD0.015L0.985LA
Tm (°C)
170
PD0.03L0.97LA
160
150
PD0.06L0.94LA
140
130
120
110
100
110
120
130
140
Tc (°C)
150
160
170
FIGURE 16.11
Melting point as a function of crystallization temperature for a series of poly(L-lactide-co-Dlactides. (From Baratian, S. et al., Macromolecules, 34, 4857, 2001. With permission.)
intersecting with the line Tm Tc), the equilibrium melting point for PLLA
has been observed to be 211°–212°C.73 However, some have questioned the
appropriateness of the Hoffman-Weeks approach for copolymers. Baratian
et al.69 elected to use a Gibbs-Thompson approach for estimating Tm0 that
requires direct measure of the lamellar thickness and uses the following
relationship:
1–2σe
Tm Tm0 ∆Hm0(lc)
冤
冥
(16.3)
where Tm is the observed melting point for lamella of thickness lc, σe is the
end (fold) surface free energy, and Hm0 is the enthalpy of fusion for 100%
crystalline PLA. Tm0 is estimated by extrapolation of Tm vs. 1/lc data to
1/lc 0. The lamellar thickness for PLLA measured by Baratian et al. agrees
reasonably well with that of Abe et al.68 Their combined data are shown
in Figure 16.12. Experimental equilibrium melting points for a series of
L-lactide-rich polymers are shown in Figure 16.13.
During normal processing and crystallization procedures, polymers do
not reach their equilibrium melting temperature. A practical equation
describing melting point for PLAs of predominantly L-lactide with mesolactide is simply Tm (°C) ≈ 175°C-300 wmeso where wmeso is less than 0.18.32 The
melting point is depressed about 3°C for every 1% initial meso-lactide.
16.5.3.2 Degree of Crystallinity
Most commonly, an enthalpy of fusion of 93.1 J/g is used for a 100% cryso
talline PLLA or PDLA homopolymers (i.e. ∆Hm
93.1 J/g)74 having infinite
1741_C16.qxd
2/11/2005
9:57 AM
Page 551
Polylactic Acid Technology
551
24
Lamellar thickness (nm)
22
20
18
PLLA (Abe)
16
14
PLLA (Runt)
12
10
110
120
130
140
Tc (°C)
150
160
170
FIGURE 16.12
PLLA lamella thickness as a function of crystallization temperature.
220
200
180
Tm
°
160
Runt - D-lactide
Runt - meso-lactide
140
Tsuji - PLLA
Runt - PLLA
Linear fit to all
120
100
0
1
2
3
4
% D isomer
5
6
7
FIGURE 16.13
Equilibrium melting point for PLA L-lactide-rich homopolymers as a function of increasing
percent D-isomer.
crystal thickness, and Equation (16.4) is used to calculate percent crystallinity from DSC scans:
∆Hm∆Hc
Crystallinity (%) 100
93.1
(16.4)
1741_C16.qxd
2/11/2005
9:57 AM
552
Page 552
Natural Fibers, Biopolymers, and Biocomposites
where ∆Hm is the measured heat of fusion and ∆Hc is the heat of crystallization. This value of 93.1 J/g is used throughout the PLA literature. However,
0
since Tm
decreases with decreasing OP, it is expected that using a constant
value for ∆Hmo across all optical compositions will introduce error. No reference was found that systematically calculated ∆Hmo for homopolymers of
reduced OP. It is well known that for PLAs of essentially random optical
copolymers of predominantly L-lactide, with small amounts of D- and mesolactides, the attainable percentage of crystallinity (xc) decreases with decreasing optical purity (OP), and crystallization is essentially nonexistent about
when OP 0.78.
16.5.3.3 Crystallization Kinetics
The crystallization kinetics of PLA are strongly dependent on the optical
copolymer composition. The degree of crystallinity, nucleation rate, and
spherulite growth rate decrease substantially with decreasing optical purity
(OP). Under quiescent crystallization conditions, Kolstad showed that the
bulk crystallization half time increases by roughly 40% for every 1 wt%
increase in the meso-lactide.75 PLA has the highest rate of crystallization
between 100° and 130°C.
Two phenomena contribute to the bulk crystallization rate, nucleation and
crystal growth. For instantaneous nucleation and spherulitic growth
(Avrami n 3), the crystallization rate constant is given by
4
k πNG3
3
(16.5)
x 1ekt
(16.6)
For the rate equation,
n
where x ∆Hm(t)/∆Hm(∞), N the density of primary nucleation sites to
initiate spherulitic growth, and G is the lineal growth rate of a spherulite.
The rate of PLA lineal crystal growth (G) of a spherulite is dependent on
Tc, OP, and molecular weight. Nulceation density (N) depends on Tc, polymer characteristics, thermal history, and impurities.
When PLA is isothermally crystallized from the melt or from a quenched
state, the size of the spherulites changes dramatically. Pluta and Galeski76
showed that spherulites were larger in samples crystallized via cooling from
the melt than those in samples crystallized via heating from the glassy amorphous state. On the lamellar level, the structures were similar for samples
prepared by the two methods under similar time and temperature conditions. The thermal properties, Tm and Hm, were similar.
Technology to increase the rate of crystallization of PLA has been investigated because even the fastest crystallizing PLAs are considered to be slow
when compared to many conventional thermoplastics. Kolstad75 showed
that talc is an extremely effective nucleating agent for PLA homopolymers.
1741_C16.qxd
2/11/2005
9:57 AM
Polylactic Acid Technology
Page 553
553
Schmidt and Hillmyer77 showed that a small level of stereocomplex crystal
is effective for seeding the crystallization of PLA homopolymers.
Liao et al.78 studied the crystallization characteristics of blends of PLLA
and PDLLA. Crystallization behavior of PLLA in the blends was similar to
that of the homopolymer when PDLLA 30%.
16.5.3.4 Stereocomplex
Stereocomplexation has been reported in the literature for a number of
polymers including polylactide. The PLA racemic crystal structure consists
of alternating L- and D-PLA chains packed side by side with a 1:1 ratio of L:D
monomer units. The chain conformation in the stereocrystal is a 31 helix vs.
the 103 for the unstrained homopolymer crystal. The 31 helix winds a little
tighter, going from 108° to 120º rotation per residue, resulting in a significant
change in the molecular shape. The 31 configuration allows a chain-to-chain
separation difference of 0.4 Å smaller.79 This tighter packing results in more
favorable nonbonding interactions that contribute to the higher crystalline
melting point. The equilibrium melting temperature of PLA stereocomplex
has been reported by Tsuji and Ikada to be 279°C,80 a very marked improvement over homocrystals, with the minimum tacticity sequencing needed to
produce stereocomplex for the L-rich and D-rich PLAs to be 15 lactyl (or 7.5
lactide) units.81 The heat of fusion for pure stereocomplex has been determined to be 142 J/g by Loomis and Murdoch.82
Melting points for stereocomplex are dictated by the optical purity (OP) of
both the L-lactide-rich and D-lactide-rich polymers (in terms of optically pure
run lengths) as well as molecular weight and blend ratio. Maximum regularity is attained with a blend of PLLA and PDLA. The greater the deviation
in OP of the L-rich and/or D-rich chains, the greater the decrease in stereocomplex melting peak. This follows the equivalent behaviors of PLA
homocrystals, but is shifted to a higher temperature. A practical upper
attainable melting temperature that would be achievable for commercial
resins approaches 230ºC.
Melting point (Tm) and enthalpy of melting (∆Hm) of the stereocomplex is
dependent on the ratio of L-lactide-rich polymer and D-lactide-rich polymer.
For a constant optical purity of racemic polymers, the Tm varies only by a
few degrees across the possible blend ratios, whereas the stereocomplex ∆Hm
varies directly with the blend ratio with maximum at 1:1. Tsuji et al. published an excellent series of papers on the crystallization, mechanical behavior, and hydrolysis properties of PLA stereocomplexes.83–94
Kang et al.95 have examined the PLA stereocomplex by Raman spectra and
determined that CO stretching bands are related to the packing formation
of the polymer chain. By combining Raman spectroscopy and DSC techniques, they deduced the enthalpy of fusion for stereocomplex to be 126 J/g.
During crystallization, packing order led the conformation order during
early stages. The CO dipole interaction is believed to be the driving force
for crystallization.
1741_C16.qxd
554
2/11/2005
9:57 AM
Page 554
Natural Fibers, Biopolymers, and Biocomposites
16.5.3.5 Stress-Induced Crystallization
While the slow quiescent crystallization kinetics of PLA may limit its speed
of processing for some applications, opportunities utilizing stress-induced
crystallization (SIC) with PLA have been enjoying success commercially
because of the naturally wide processing window. It is possible to highly orient PLA in the rubbery or molten state during operations such as film
stretching or fiber spinning. Fabrication processes such as this take full
advantage of the semicrystalline nature of PLA to develop different properties. Kokturk et al.96 showed that PLA exhibited almost ideal stress-strain
behavior in the temperature range 65°–80°C. As films are stretched, crystallization and strain hardening leads to uniform thickness at high deformation. This effect is referred to as self-leveling, and it is a highly desirable
property for orientation processes. Crystallization of homopolymers, during
orientation processes, results in β crystals in a 31 helix conformation.
Puiggali et al.97 characterized PLA crystals made through SIC, using electron diffraction and conformational energy analysis. Chains were shown to
be arranged as a frustrated packing of 31 helixes in a trigonal unit-cell of
parameters a b 1.052 nm, c 0.88 nm. The frustrated packing is of the
type described as North-South-South (NSS).
The effects of uniaxial drawing conditions on the orientation of various
poly(L-lactide) (PLLA) samples have been investigated using Fourier transform IR dichroism and thermal analysis.98 Orientation in uniaxially stretched
PLLA was influenced by the drawing rate, drawing temperature, and draw
ratio. It was shown that spherulites could be deformed at high temperature
under drawing conditions into new crystal morphologies.
Kister et al.99 first examined the Raman and IR spectra of poly(L-lactic
acid) in the amorphous and semicrystalline states from 3600 to 100 cm1.
Vibrational assignments were proposed, and Kister et al. indicated that
Raman data supported the 103 helical conformation for PLA. Raman spectroscopy has been further developed for PLA more recently. Smith et al. and
Kang et al. have reported on a technique for characterizing orientation in
both the amorphous and crystalline phases for films in multiple axes.100–102
Correlations to level of crystallinity were also developed. Film shrinkage
was correlated to crystallinity level and orientation of the amorphous phase.
Randall et al.103 studied the melting points of PLAs made by SIC fabrication processes of very different time scales using DSC. At a given OP, PLA
homopolymers tended to develop the same melting point during film orientation, bottle inflation, and fiber spinning.
Cicero et al.104,105 characterized the crystal morphology of fibers made with
linear and branched architectures. The level of crystallinity increased with
draw ratio. The morphology was fibrillar, with microfibril diameters ranging
from 20 to 30 nm for both architectures. Cicero et al. noted that the orientation of the crystal blocks was similar to nylon. Two-step drawn fibers
undergo a morphology transition from class 2 to class 1 within the classification proposed by Keller et al.106 at higher draw conditions.
1741_C16.qxd
2/11/2005
9:57 AM
Page 555
Polylactic Acid Technology
16.5.4
555
Degradation and Hydrolysis
Under conditions of high temperature and high humidity, as in active compost, for example, PLA will degrade quickly and disintegrate within weeks
to months. The primary mechanism of degradation is hydrolysis, followed
by bacterial attack on the fragmented residues. The environmental degradation of PLA occurs by a two-step process.107 During the initial phases of
degradation, the high molecular weight polyester chains hydrolyze to lower
molecular weight oligomers. The rate of hydrolysis is accelerated by acids or
bases and is dependent on moisture content and temperature. Article dimensions, crystallinity, and blends will affect the rate of degradation. PLA products rapidly degrade in both aerobic and anaerobic composting conditions:
Stage
Molecular
Weight
Rate of MW
Decrease
First stage
High
Second stage
Critical (Mn: 10,000
– 20,000)
Low
Weight
Loss
Slow (rate
None
determined)
Onset
Rapid
Rapid
Hydrolysis
Reaction
Nonenzymatic
Degradation
Mechanism
Bulk
Nonenzymatic
Bulk and
and enzymatic
surface
Under typical use conditions, PLA is very stable and will retain its molecular weight and physical properties for years. This is typified by its growing
use in clothing and durable applications. High molecular weight PLA is also
naturally resistant to supporting bacterial and fungal growth, which allows
it to be safely used for applications such as food packaging and sanitation.
A recent review by Tsuji108 gives an excellent account of PLA degradation
studies.
16.5.4.1 Hydrolysis
Degradation of PLA is primarily due to hydrolysis of the ester linkages,
which occurs more or less randomly along the backbone of the polymer. It
requires the presence of water according to the following reaction:
CH 3
CH 3
PLA
CH
C
O
CH
O
CH
OPLA
+ H2O
O
CH 3
PLA
C
CH 3
C
O
OH + HO
CH
C
O
OPLA
1741_C16.qxd
2/11/2005
9:57 AM
Page 556
556
Natural Fibers, Biopolymers, and Biocomposites
The rate of hydrolysis is determined by its intrinsic rate constant, water
concentration, acid or base catalyst, temperature, and morphology. Two
major challenges to the stabilization of PLA with regard to hydrolysis are the
fact that it is quite permeable to water and that the hydrolysis reaction is
autocatalytic. Polyethylene terephthalate, for example, is also a polyester
that hydrolyzes, but the inherent rate of hydrolysis is slower than PLA and
it is not autocatalytic.
The autocatalytic hydrolysis reaction is the following:
R-COOH
ka
R-COO¯ + H +
CH 3
CH 3
PLA
CH
C
O
O
CH
C
OPLA + H 2 O + H +
kh
O
CH 3
PLA
CH
CH 3
C
OH + HO
CH
O
C
OPLA + H +
O
An equation describing decrease in ester concentration [E] over time is the
following:
d(1/Mn)
d[E]
k[COOH][H2O] dt
dt
(16.7)
for random chain scission where [COOH] ∝ 1/Mn and the product
[H2O][E] is constant. Rearranging to
Mnd(1/Mn)kdt
(16.8)
ln Mn,t lnMn,okt
(16.9)
the integrated form becomes
where Mn,t number average molecular weight at time t, Mn,o number average molecular weight at time zero, and k is the hydrolysis rate constant. These
kinetics were derived by Pitt et al.,109 and were again supported by Tsuji.108
Siparsky et al.110 proposed a slightly different equation using an autocatalytic model:
d[COOH]
kh[COOH][H2O][E]
dt
(16.10)
They found that this equation did not fit their solution kinetics so they
derived a second equation using quadratic autocatalysis kinetics. This model
1741_C16.qxd
2/11/2005
9:57 AM
Page 557
Polylactic Acid Technology
557
fit the data much better. It is of the following form:
d[x]
kh[E][H2O][x]
dt
where x Ka[COOH]1/2, and Ka is the acid dissociation constant. This form
accounts for the dissociation of the acid.
Stated differently,
d[E]
kh[H2O][E]([COOH]ka)1/2
dt
(16.11)
The expression makes it possible to take into account all the acid species in
the polymer. It also removes the water dependence of the rate constant
expression. A comprehensive predictive hydrolysis model for PLA autocatalysis also requires temperature dependence for kh above and below Tg,
and accurate prediction of water concentration. Combining this expression
with mass transfer terms for [H2O] and [COOH] as a function of position
in an article may be useful for some applications.
Strategies for stabilizing PLA to hydrolysis include reducing the level of
residual monomer to as low a level as possible and lowering the water concentration in PLA and preventing autocatalysis. The equilibrium moisture
content can be decreased by controlling morphology (highly crystalline, oriented, fibular structure). Reduction in the rate of autocatalysis has been
accomplished by incorporation of basic, buffering salts such as CaCO3.
Certain impurities and classes of additives increase the rate of PLA hydrolysis, including lactide and oligomers and some acidic and basic additives. A
second approach to preventing autocatalysis was reported by Lee et al.111 by
functionalizing the PLA end group chemistry. End groups studied included
OH-, COOH-, Cl-, and NH2-. NH2- and Cl-terminated PLAs were more
resistant to thermal and hydrolytic degradation. The thermal stability of the
OH-terminated PLA was poor (likely due to lactide formation), and
the hydrolytic stability of the COOH-terminated PLA was also poor.
The hydrolysis of PLA has been studied using at least four experimental
designs: melt hydrolysis (discussed in rheology section above) to simulate
extrusion conditions, solution hydrolysis of PLA dissolved in various solvents, and two types of solid-state hydrolysis methods. The first type of
solid-state hydrolysis experiment utilizes solid samples suspended in aqueous media. The second type utilizes exposure of solid samples to humidity.
Since the product of the hydrolysis reaction is also a catalyst for further reaction, the hydrolysis of PLA is autocatalytic. Melt hydrolysis, solution hydrolysis, and solid exposure to humidity experimental designs generally give
rise to homogeneous hydrolysis. In contrast, exposure of solid samples suspended in water or aqueous buffers gives rise to very inhomogeneous
hydrolysis. This is due to differences in hydrolysis rate as a function of depth
within the sample and it is caused by extraction of PLA oligomers from the
surface of the sample into the aqueous medium. It is the chain end groups of
the PLA oligomers that catalyze the rate of hydrolysis, and therefore the
1741_C16.qxd
558
2/11/2005
9:57 AM
Page 558
Natural Fibers, Biopolymers, and Biocomposites
inside of the sample actually hydrolyzes faster than the surface for thick
samples submersed in water.
It is important that the data used for the determination of any type of stability analysis be consistent with the type of hydrolysis conditions of the
intended application. In vivo degradation of PLA is outside the scope of this
chapter.
Four key features mark the hydrolytic degradation of PLA:
1. The pKa of PLA’s carboxylic acid end group and its oligomers is
unusually low (~3) compared to most carboxylic acid groups (4.5
to 5). PLA’s carboxylic acid end groups catalyze the ester hydrolysis, which generally leads to a faster rate of degradation as the
polymer degrades (autocatalysis).
2. Two reaction mechanisms apparently occur. The first is a random
chain scission and the second a chain-end hydrolysis. The chainend scission is about 10 times faster than the random scission reaction at low pH values but is negligible at neutral and basic pH
values. The number of ester groups along the chain is much greater
than that of the chain ends and is statistically much more important.
3. The crystalline regions hydrolyze much more slowly than the
amorphous regions.
4. The rate of hydrolysis (slope) is much greater above the glass transition temperature (Tg) than below.
In the following, we summarize each of the four hydrolysis experimental
design conditions independently. Melt hydrolysis is covered in the rheology
section pertaining to melt stability.
16.5.4.2 Hydrolysis of Solid Samples Suspended in Aqueous Media
Hydrolytic degradation of PLA is autocatalytic due to the production of carboxylic acid end groups that catalyze further hydrolysis.112 An effect of this
is that for a thick sample immersed in a 7.4 pH buffer at 37°C, bulk hydrolysis occurs faster than surface hydrolysis.113,114 The pH of the surface remains
that of the 7.4 pH buffer whereas the pH of the interior of the sample is
reduced due to the accumulation of PLA acid end groups and the inability
of the oligomers to quickly diffuse into the buffer medium.
Vert et al.115 summarized the literature on the hydrolytic stability of solid,
suspended PLA in 1997 and captured several key points including autocatalysis and leaching effects. In addition, morphology and structure were
addressed. The rate of degradation is strongly influenced by the number
average molecular weight.116 Lower Mn PLA degraded faster as did PLA
with higher levels of oligomers. The presence of lactide also increased the
rate of hydrolytic degradation.117 Effects of Mn, oligomers, and lactide can all
be attributed to autocatalysis. PLA’s amorphous domains are more susceptible to hydrolysis than the crystalline domains. This is due to the ability of
1741_C16.qxd
2/11/2005
9:57 AM
Polylactic Acid Technology
Page 559
559
water to permeate within the amorphous phase but not the crystalline phase,
and leads to differences in water concentration throughout the article on the
microscale. Hydrolysis of the crystalline domain occurs mainly by a surface
erosion mechanism.
Grizzi et al. used the fact that surface hydrolysis is slow compared to bulk
hydrolysis to stabilize PLA to hydrolysis simply by using small dimensions.113 By suspending PLA samples of varying size in water and determining the rates of hydrolysis, they showed that leaching could be used to slow
down the rate of solid-state hydrolysis. They estimated that the skin, which
was formed due to the leaching of catalytic oligomers, was about 200 to
300 µm in thickness. Samples smaller than this size were found to degrade
much more slowly than large samples.
The heterogeneity of the process makes it difficult to determine hydrolysis
kinetics constants for solids suspended in aqueous media. Translation of data
from these kinds of studies to other environments is suspect for this reason.
In alkaline media, Iwata and Doi118 studied the change in morphology and
crystalline state of solution-grown single crystals of PLLA. Samples were
characterized by transmission electron microscopy (TEM), atomic force
microscopy (AFM), and gel permeation chromatography (GPC). As hydrolysis proceeded, the molecular weight of the degraded crystals corresponded
to the value calculated from lamellar thickness measured by AFM. The
explanation for this is that the PLA first degrades in the more loosely packed
chain folding regions. In the later stages of hydrolysis, the more persistent
crystalline region’s molecular weight approaches that of the lamellar thickness, and the degradation mechanism will change to that of surface erosion.
Tsuji and Nakahara119 tested PLLA films at pH 2.0 (1. HCl solution,
2. D,L-lactic acid solution), 7.4 (phosphate buffered), and 12 (NaOH soln).
While the alkaline media degraded faster than the neutral pH control, the
samples in acid media degraded at about the same rate as the neutral pH control. Using microcapsules suspended in buffer solutions, Makino et al.120
showed that PLA degrades slowest at pH 5.0 and increases in acidic and alkaline solutions. The activation energy of hydrolysis at pH 7.4 and ionic strength
0.15 was 19.9 kcal/mol for PDLLA microcapsules and 20.0 kcal/mol for PLLA
microcapsules when the initial Mw was about 100,000 for both polymers.
The degradation characteristics of PLA in seawater are important for certain applications and potential disposal routes. Yuan et al.121 tested PLLA
fibers in saline buffer at 37°C for molecular weight loss and mechanical
properties. The results suggested that the tensile properties of the PLLA
fibers were stable for 35 weeks, even though their molecular weights
decreased remarkably and obvious spotty defects appeared on their surfaces. Tsuji and Suzuyoshi122 showed that PLA degraded faster in natural
seawater than controlled static seawater, although the rate of molecular
weight loss is slow. Makino et al.120 showed that AlCl3, CaCl2, and KCl had
no or little effect on the degradation rate.
At temperatures above the glass transition, PLA’s degradation rate increases
substantially. Tsuji et al.123 tested PLLA films in buffers at 97°C and showed
1741_C16.qxd
560
2/11/2005
9:57 AM
Page 560
Natural Fibers, Biopolymers, and Biocomposites
that the hydrolysis takes place predominantly and randomly at the chains in
the amorphous region. Amorphous samples crystallized while degrading and
higher initial crystallinity slightly slowed the rate of hydrolysis.
In a very extreme case showing the effect of morphology, fibers (drawn to
a draw ratio of about 8 using a solution spinning approach) were examined. These fibers possessed few defects (chain folds, microvoids) together
with high crystallinity, low fractional free volume, and small cross-sectional
dimension.112 The PLLA fiber was very stable over a 5-year period immersed
in water at room temperature under a static load and lost 10% of its diameter and less than 25% of its tensile strength. The perfection of morphology
was thought to reduce the penetration and diffusion of water into the fiber.
Effects of processing conditions were also examined by Iannace et al.124
Samples of PLLA crystallized under isothermal and nonisothermal conditions were prepared. Initial morphology developed during the isothermal
crystallization affected the rate of hydrolysis in buffer solutions. Larger crystals developed at higher Tc were more resistant to erosion compared to less
perfect and smaller crystals developed at higher degrees of supercooling.
Tsuji and Suzuki94 also studied stereocomplex degradation over the course
of 30 months, in 7.4 pH buffer solution using 1:1 blends and nonblended
films prepared from PLLA and PDLA. Properties and molecular weight
were monitored with GPC, tensile tests, DSC, SEM, optical polarizing
microscopy, x-ray diffractometry, and gravimetry. The rate of molecular
weight loss, tensile strength, Young’s modulus, melting temperature, and
mass remaining of the films in the course of hydrolysis was more stable for
the stereocomplex film than for the nonblended films.
16.5.4.3 Solution Hydrolysis
Solution hydrolysis is generally thought to occur by random scission of the
ester linkages in PLA.125 It has been reported, however, that base and neutral
hydrolysis occur by random chain scission but acid hydrolysis occurs by a
mixture of two mechanisms, random and chain-end scission, with chain-end
scission being the faster.126 This was determined in dioxane/water solution
with 1H NMR and GPC using a statistical method where the level of lactic
acid produced by hydrolysis is compared to the degree of degradation
across the entire molecular weight distribution. The chain-end scission is
sometimes referred to as “unzipping.”
PLA solution hydrolysis was also studied in acetone/water by Zhang
et al.127 and acetonitrile/water by Siparsky et al.110 The latter solvent had a
much higher dielectric constant (ε 36) than acetone (ε 21) or dioxane (ε 2)
so that the carboxylic acid groups were more ionized, as in water (ε 79).
Autocatalysis was observed in acetonitrile whereas it was not in the lower
dielectric solvents due to a lower extent of ionization of the carboxylic acid.
The solution hydrolysis rate constant (in dioxane solution) did not depend
on the optical purity (OP) of the polymer whereas it does in solid-state
degradation (probably due to crystallinity differences).128 CaCO3 was
1741_C16.qxd
2/11/2005
9:57 AM
Page 561
Polylactic Acid Technology
561
observed to slow the rate of hydrolysis due to its buffering capacity and
water insolubility.
16.5.4.4 Hydrolysis of Samples Exposed to Humidity
Samples exposed to temperature and humidity best simulate the shelf-life
environment for many PLA applications. The equilibrium moisture level as
a function of humidity is known for PLA at room temperature. It has not
been as thoroughly studied for elevated temperatures. Hydrophilic impurities such as lactide and carboxylic acid end groups, whose concentration
increase as degradation proceeds, increase the level of moisture.
Sharp et al.129 measured the water uptake of PDLLA using a quartz crystal microbalance in humid environments at various temperatures. Kinetics of
water adsorption and equilibrium concentrations were measured, and the
Flory Huggins interaction parameter X was found to be 3.5 at 20°C using
binary mixing theory. The equilibrium swell level of water in PLA increases
with decreasing molecular weight due to the strong affinity of acid and
hydroxyl end groups for water. It was also shown that the Tg of high molecular weight PLA was depressed by 11°C at 7000 ppm water (the highest level
of swelling studied) compared to zero ppm water.
Witzke32 measured the swelling concentrations of water in PLA using
Karl-Fisher coulometric titration and arrived at an empirical model fit for
estimating swell concentration:
Water (ppm) 40.8RH(%)6.6107RH(%)5
(16.12)
where water is in parts per million by weight and RH is the relative humidity. A plot of this equation is shown in Figure 16.14.
Moisture content (ppm)
Low
16000
15000
14000
13000
12000
11000
10000
9000
8000
7000
6000
5000
4000
3000
2000
1000
0
0
10
Typical
20
30
40
50
60
Humidity (%RH)
High
70
80
90
FIGURE 16.14
Equilibrium swelling concentration of water in high molecular weight PLA in humid
environments.
100
1741_C16.qxd
2/11/2005
9:57 AM
Page 562
562
Natural Fibers, Biopolymers, and Biocomposites
Crystalline pellets reach an equilibrium level of moisture much more slowly
(by a factor of 2.4) than amorphous pellets, and crystalline pellets dry 1.5 times
slower than amorphous pellets130 due to differences in water diffusion.113
However, Siparsky et al. found that once PLA was equilibrated with moisture,
crystallinity had little effect on the solubility or diffusion constant of water.131
They also showed that the diffusion of water in PLA is much faster than the
rate of hydrolysis in the temperature range between 20° and 50°C.
Li and McCarthy found that the water absorption of PLA increased
markedly as the hydrolysis proceeded, presumably due to the production of
the more hydrophilic acid and alcohol end groups.132 Others substantiated
this point.133 Two distinct populations of absorbed water were observed for
poly(D,L-lactide-co-glycolide), one bound and one mobile.134 The rate of
hydrolysis of the copolymer correlated with the level of bound water.
The rate of hydrolysis increases markedly above Tg. Compost studies on
PLA articles show that at temperatures above the Tg, the rate of hydrolysis is
fast enough to completely consume articles in a matter of months. Results of
a Cargill Dow LLC internal study showing Mw loss over time at various
aggressive storage conditions are shown in Figure 16.15.
Lactide has a large effect on hydrolytic degradation in humid environments because it rapidly hydrolyzes to a relatively strong acid, which promotes hydrolysis. At high levels (2%) the presence of lactide has a dramatic
effect on PLA molecular weight loss. Acidic or basic additives catalyze the
rate of PLA hydrolysis. MgSO4 and SiO2 promoted degradation whereas
CaCO3 inhibited hydrolytic degradation.135 Li and Vert showed that the presence of porous coral in a blend with PLA (60% coral) dramatically reduced
the rate of PLA degradation by neutralizing the effect of autocatalysis.136
1000000
Mw by GPC
40°C, 50% RH
40°C, 80% RH
100000
10000
60°C, 80% RH
80°C, 80% RH
80°C, 20% RH
1000
0
20
40
60
80
100 120
Time (days)
140
160
180
200
FIGURE 16.15
Cargill Dow LLC internal study showing molecular weight loss for amorphous PLA stored at
various aggressive temperatures and humidity.
1741_C16.qxd
2/11/2005
9:57 AM
Page 563
Polylactic Acid Technology
563
The crystalline regions of PLA were found to hydrolyze very slowly.112,125
The rate of molecular weight loss follows a two-step process, the first occurring very fast (interpreted as loss of the amorphous phase) and the second
occurring much slowly (interpreted as loss of the crystalline phase). The
apparent percentage of crystallinity increases during the amorphous
phase hydrolysis, consistent with the interpretation. The molecular weight
of the remaining PLA after the amorphous phase was degraded was about
5000 to 9000 (DP 70 to 125), roughly that of the fold length of the
lamella.74,137 Crystalline injection-molded samples lost physical properties
faster than amorphous samples, although the rate of Mw loss was similar
between them.125 This was interpreted as the loss of amorphous tie chains
between the crystals causing catastrophic failure of the system. Also, chain
folds and tie chains connecting crystallites are thought to hydrolyze preferentially because they are stressed due to a “reeling in” force.138 These observations may be the result of chain end groups and catalysts concentrating in
the amorphous regions.
Morphology also affects water permeation in another way. Microvoids or
porosity allow water to migrate more freely into the material and thereby
enhance the rate of hydrolysis. This is probably a small effect for samples
exposed to humidity because the rate of hydrolysis is slow compared to
the water diffusion rate, as discussed above. Localized regions of high
water absorption may cause mechanical stresses due to increased osmotic
pressure.
16.5.5
Enzymatic Degradation
Several enzymes have been shown to catalyze PLA hydrolysis. Enzymatic
hydrolysis of PLLA with varied crystal morphologies was studied using proteinase K by Tsuji and Miyauchi.139,140 It was concluded that the enzymatic
degradation proceeded via mainly a surface erosion mechanism. Amorphous
regions hydrolyzed more readily than crystalline regions. The enzymatic
hydrolysis rate (REH) was modeled using the following equation:
REH(µg/(mm2·h)0.36(1xc)
(16.13)
where xc the fraction of crystalline PLLA. Tie chains degraded more readily than folds or restricted amorphous chains.
16.6
Applications and Performance
Since PLA is an environmentally friendly polymer that can be designed to
controllably biodegrade, it is ideally suited for many applications in the environment where recovery of the product is not practical, such as agricultural
1741_C16.qxd
2/11/2005
9:57 AM
Page 564
564
Natural Fibers, Biopolymers, and Biocomposites
mulch films and bags. Composting of post-consumer PLA items is also a
viable solution for many PLA products. However, the large growth seen for
PLA in many applications does not depend upon the biodegradability of the
material.
PLA resins can be tailor-made for different fabrication processes, including injection molding, sheet extrusion, blow molding, thermoforming,
film forming, or fiber spinning. The key is controlling certain molecular
parameters in the process such as branching, D-isomer content, and molecular weight distribution. The ability to selectively incorporate L-, D-, or
meso-lactide stereoisomers into the polymer backbone allows PLA to be
tailored for specific applications. The ease of incorporation of various
defects into PLA allows for control of both crystallization rate and ultimate
crystallinity.
Typical properties of NatureWorks PLA from Cargill Dow LLC for injection molding and extrusion applications are given in Table 16.6. The various
grades of NatureWorks PLA differ in stereochemical purity, molecular
weight, and additive packages. Each grade is optimized for both processing
and end use performance in its intended application.
TABLE 16.6
Properties of Extrusion/Thermoforming and Injection Molding Grades of
NatureWorks PLA
PLA Polymer
2000Da
Physical Property
Specific gravity (g/cc)
Melt index, g/10 min
(190°C/2.16 kg)
Clarity
Mechanical Properties
Tensile strength at break,
psi (MPa)
Tensile yield strength,
psi (MPa)
Tensile modulus,
kpsi (GPa)
Tensile elongation (%)
Notched izod impact,
ft-lb/in (J/m)
Flexural strength,
psi (MPa)
Flexural modulus,
kpsi (GPa)
a
b
1.25
4–8
ASTM
Method
D792
D1238
Transparent
PLA Polymer
3010Db
1.21
10–30
ASTM
Method
D792
D1238
Transparent
7,700 (53)
D882
8,700 (60)
D882
500 (3.5)
D882
6.0
0.24 (0.33)
D882
D256
7,000 (48)
D638
2.5
0.3 (0.16)
D638
D256
12,000 (83)
D790
555 (3.8)
D790
2000D is a product of Cargill Dow LLC designed as an extrusion/thermoforming grade;
properties typical of extruded sheet.
3010D is a product of Cargill Dow LLC designed as an injection molding grade; properties
typical of injection molded tensile bars.
1741_C16.qxd
2/11/2005
9:57 AM
Page 565
Polylactic Acid Technology
565
Injection molding of heat-resistant products requires rapid crystallization rates that can be achieved by PLA, typically containing less than 1%
D-isomer and often with the addition of nucleating agents.75 These compositions allow high levels of crystallinity to develop during the fast cooling
cycle in the mold. Extrusion-thermoforming is optimized at a D-isomer content that does not allow crystallization to occur during the melt processing
steps, with 4–8% D-content being the effective range. Branching can be introduced by a variety of methods,62–64, 66 thus enhancing melt strength during
fabrication and opening up new application opportunities in the areas of
foams and extrusion coating. Examples of a number of articles fabricated
from PLA are shown in Figure 16.16. In short, the rheological characteristics
and physical properties of PLA can be tailored for use in a variety of
processes and applications.
Fiber is one of the largest potential application areas for PLA. PLA is readily melt spinnable, stress crystallizes upon drawing, and can be designed for
FIGURE 16.16
Articles produced from PLA. (a) Injection stretch-blow molded bottles. (b) Films. (c)
Extrusion-thermoformed containers. (d) Fiberfill products. (e) Carpet and coverings.
512
2.9
5.75
26.2
Denier (dpf)
Total denier/filament count
Strength (kg)
Tenacity (g/den)
Elongation (%)
d
c
b
a
21
61
2.8
4.75
114/24
10
29
5.7
3.04
73/24
FDYd
6700D is a product of Cargill Dow LLC designed for extrusion into mechanically drawn monofilament.
6200D is a product of Cargill Dow LLC designed for extrusion into mechanically drawn staple fibers or continuous filaments.
Partially Oriented Yarn (POY).
Fully Drawn Yarn (FDY).
0
35
5.7
1.5
staple/N.A.
POYc
D1907
D1907
D2256
D2256
multifilament,
D3822 single
filament
D2256
multifilament,
D3822 single
filament
D2102
D792
D1238
D3417
D3418
D1238A & B
566
Boiling water shrinkage (%)
Cotton-Type
Staple
1.25–1.28
1.08–1.12
60–65 (140–150)
160–170 (320–340)
15–30
1.25–1.28
1.08–1.12
55–60 (130–140)
145–155 (295–310)
5–15
ASTM Method
9:57 AM
Monofilament
PLA Polymer 6200Db
PLA Polymer 6700Da
2/11/2005
Fiber Properties
Specific gravity
Melt density, 230°C (445°F)
Tg, °C (°F)
Tm, °C (°F)
Melt flow rate, 210°C (410°F)
Material Properties
Properties of NatureWorks PLA Fabricated into Fibers
TABLE 16.7
1741_C16.qxd
Page 566
Natural Fibers, Biopolymers, and Biocomposites
1741_C16.qxd
2/11/2005
9:58 AM
Page 567
Polylactic Acid Technology
567
many fiber applications. Some of the current fiber uses include hollow fiberfill for pillows and comforters, bulk continuous filament (BCF) for carpet, filament yarns, and spun yarns for apparel, spunbond, and other nonwovens
and bicomponent fibers for binders and self-crimping fibers. Most fiber
applications require polymer with high optical purity (OP) to allow high levels of crystallinity to develop and to have adequate heat resistance in the
application. Binder fibers are unique in that low crystallinity in the sheath
layer is desired to allow ease of melting and adhesion to other fibers; thus,
high (8–20%) D- or meso-lactide content is incorporated. PLA can be
processed on standard thermoplastic fiber spinning equipment with the
appropriate temperature profiles relative to its crystal melting point. Melt
temperatures of 200°–240°C are typically used. As with all melt processing
of PLA, care must be taken to ensure that the material is dry and does not
pick up moisture, otherwise unacceptable molecular weight loss will occur.
Typical drying conditions are 2–4 h in a hopper dryer with – 40oC dew point
air with a resulting moisture content of less than 50 ppm. Some of the properties of PLA fibers are shown in Table 16.7 and compared to other natural
and synthetic fibers in Table 16.8.
PLA fiber can be combined with natural or regenerated fibers including
cotton, wool, silk, viscose, lyocell, and others along with synthetic fibers
made from PET, nylon, and other petroleum-based synthetics. PLA can be
included as a minor component (5–15%) or as the major fiber, depending on
the balance of properties and appearance desired. One of the fastest growing
application areas is in nonwoven wipes containing 35% PLA and 65% viscose.
PLA is replacing PET in these applications because of its superior performance and the fact that the disposable products can be produced from fibers
that are from 100% renewable resources and are also 100% biodegradable.
Some of the beneficial characteristics of PLA fiber products include its natural soft feel, ease of processing, and unique stain and soil resistance. PLA
excels at resistance to stain in standard tests with coffee, cola, tea, catsup, lipstick, and mustard. PLA also burns with low smoke generation, has good
ultraviolet resistance, is easily dyeable, and brings good wickability of moisture to applications.
TABLE 16.8
Properties of Synthetic and Natural Fibers
Fiber Property
Nylon 6
PET
Acrylics
PLA
Rayon Cotton Silk
Specific gravity
1.14
1.39
1.18
1.25
1.52
Tenacity (g/d)
5.5
6.0
4.0
6.0
2.5
Moisture regain (%)
4.1
0.2–0.4
1.0–2.0
0.4–0.6 11
Elastic recovery
89
65
50
93
32
(5% strain)
Flammability
Medium High
Medium Low
Burns
smoke
smoke
UV resistance
Poor
Fair
Good
Good
Poor
Wool
1.52
4.0
7.5
52
1.34
4.0
10
52
Burns
Burns Burns
slowly
Fair
Fair
Fair
1.31
1.6
14–18
69
1741_C16.qxd
2/11/2005
9:58 AM
568
Page 568
Natural Fibers, Biopolymers, and Biocomposites
TABLE 16.9
Properties of Biaxially Oriented NatureWorks PLA
Film Properties
Specific gravity (g/cc)
Tensile strength (kpsi)
MDb
TD
Tensile modulus (kpsi)
MD
TD
Elongation at break (%)
MD
TD
Elmendorf tear (g/mil)
MD
TD
Spencer impact (J)
Transmission rates
O2, cc-mil/m2/24 h atm
CO2, cc-mil/m2/24 h atm
H2O, g-mil/m2/24 h atm
Optical characteristics
Haze (%)
Gloss, 20°
Thermal characteristics
Tg, °C (°F)
Tm, °C (°F)
a
b
PLA Polymer 4040Da
ASTM Method
1.25
D1505
16
21
D882
D882
480
560
D882
D882
160
100
D882
D882
15
13
2.5
D1922
D1922
550
3000
325
D1434
D1434
E96
2.1
90
D1003
D1003
52 (126)
135 (275)
D3418
D1003
4040D is a product of Cargill Dow LLC suitable for preparation of biaxially oriented film.
Properties on 1 mil film oriented 3.5 in MD and 5 in TD.
MD refers to machine direction and TD refers to transverse direction.
Films are the second largest application area for PLA. Again, the ability to
modify the crystallization kinetics and physical properties for a broad range of
applications by D- or meso-comonomer incorporation, branching, and molecular weight change makes PLA extremely versatile. Films are transparent when
stress crystallized and have acceptance by customers for food contact. PLA
films can be prepared by the blown double bubble technology or preferably,
cast-tentering. Cast-tentered films have very low haze, excellent gloss, and gas
(O2, CO2, and H2O) transmission rates desirable for consumer food packaging.
PLA films also have superior dead fold or twist retention for twist wrap packaging. Some of the properties of PLA films are shown in Table 16.9.
16.7
Summary
Polymers of polylactic acid have finally become a commercial reality with the
construction of a world-scale plant. The cost-performance balance of PLA has
1741_C16.qxd
2/11/2005
9:58 AM
Polylactic Acid Technology
Page 569
569
resulted in its use in many applications, including packaging, paper coating,
fibers, films, and a host of molded articles. PLA is not being used in these
applications solely because of its degradability nor because it is made from
renewable resources. PLA is in both durable and short-term applications. The
use of PLA as a cost-effective alternative to commodity petrochemical-based
plastics will increase demand for agricultural products such as corn and
sugar beets, and is an advanced example of sustainable technology.
One of the additional benefits of large commercial production of lactic
acid for PLA is the increased availability and reduced cost of lactic acid or
lactide for the production of lactic acid esters. These esters, especially ethyl
lactate, are finding increased use as environmentally friendly solvents,
cleaning agents, and diluents. They are replacing more hazardous solvents
such as chlorinated organic and aromatic compounds.
As we move into the 21st century, increased utilization of renewable
resources will be one of the strong drivers for sustainable products. Reduced
energy consumption, waste generation, and emission of greenhouse gases
will take on greater emphasis. Polylactic acid is the first commodity plastic
to incorporate these principles and Cargill Dow LLC was awarded the 2002
Presidential Green Chemistry Award for their process to produce
NatureWorks PLA.141 Future developments will include blends of PLA,
copolymers, and impact-modified products, which will further expand the
applications where this unique polymer can be used.
Acknowledgments
The authors would like to thank Kevin McCarthy, Patrick Smith, and Ray
Drumright for their contributions to this manuscript, and Cargill Dow, LLC
for permission to publish it.
Reference
1. Lipinsky, E.S. and Sinclair, R.G., Chem. Eng. Prog., 82, 26–32, 1986.
2. Carothers, H., Dorough, G. L., and Van Natta, F. J., J. Am. Chem. Soc., 54, 761, 1932.
3. Nef, J. U., Liebigs. Ann. Chem., 403, 204, 1914; Pelouze, P. M. J., Ann. Chim. Phys.,
13, 257, 1845.
4. Plastics Technology, January 1998, pp. 13–15.
5. Enomoto, K., Ajioka, M., and Yamaguchi, A., U.S. Patent 5,310,865, 1995;
Kashima, T., Kameoka, T., Ajioka, M., and Yamaguchi, A., U.S. Patent 5,428,126,
1995; Ichikawa, F., Kobayashi, M., Ohta, M., Yoshida, Y., Obuchi, S., and Itoh, H.,
U.S. Patent 5,440,008, 1995; Ohta, M., Obuchi, S., and Yoshida, Y., U.S. Patent
5,440,143, 1995.
1741_C16.qxd
570
2/11/2005
9:58 AM
Page 570
Natural Fibers, Biopolymers, and Biocomposites
6. Gruber, P.R., Hall, E.S., Kolstad, J.J., Iwen, M.L., Benson, R.D., and Borchardt,
R.L., U.S. Patent 5,142,023, 1992; Gruber, P.R., Hall, E.S., Kolstad, J.J., Iwen, M.L.,
Benson, R.D., and Borchardt, R.L., U.S. Patent 5,247,058, 1992; Gruber, P.R., Hall,
E.S., Kolstad, J.J., Iwen, M.L., Benson, R.D., and Borchardt, R.L., U.S. Patent
5,247,059, 1993; Gruber, P.R., Hall, E.S., Kolstad, J.J., Iwen, M.L., Benson, R.D.,
and Borchardt, R.L., U.S. Patent 5,258,488, 1993; Gruber, P.R., Hall, E.S., Kolstad,
J.J., Iwen, M.L., Benson, R.D., and Borchardt, R.L., U.S. Patent 5,274,073, 1993;
Gruber, P.R., Hall, E.S., Kolstad, J.J., Iwen, M.L., Benson, R.D., and Borchardt,
R.L., U.S. Patent 5,357,035, 1994; Gruber, P.R., Hall, E.S., Kolstad, J.J., Iwen, M.L.,
Benson, R.D., and Borchardt, R.L., U.S. Patent 5,484,881, 1996.
7. Chem. Week, May 2000, 162, pp. 24E.
8. Hartmann, M.H., Biopolymers from Renewable Resources, Kaplan, D.L., Ed.,
Springer, Berlin, 1998, chap. 13; Dubois, P., Jerome, R., Lofgren, A., and
Albertsson, A.C., J.M.S. Rev. Macromol. Chem. Phys., 35, 379, 1995; Tsuji, H.,
Polylactides in Biopolymers, Wiley-VCH, 2002, chap. 5; Drumright, R., Gruber, P.,
and Henton, D. E., Adv. Mater., 12, 1841, 2000.
9. Holten, C.H., Lactic Acid, Properties and Chemistry of Lactic Acid and Derivatives,
Verlag Chemie, GmbH, Weinheim/Bergstr., 1971.
10. Schopmeyer, H.H., Lactic Acid, in Industrial Fermentations, Underkofler, L. and
Hickley, R.J., Eds., Chemical Publishing Co. New York, NY, 1954, chap. 12.
11. Cox, G. and Macbean, R., Lactic Acid Recovery and Purification Systems, Research
Project Series, No. 29, October, 1976.
12. Busche, R., Shimshick, E., and Yates, R., Biol. Bioeng. Symp., No. 12, 1982, pp.
249–262.
13. Gruber, P., Keynote Address, Massachusetts Green Chemistry Symposium, Cosponsored by National Environmental Technology Institute, UMASS and MA
Executive Office of Environmental Affairs Toxic Use Reduction Institute,
Amherst, MA, Oct. 30, 2001.
14. Kricheldorf, H.R. and Krieser, I., Makromol. Chem., 187, 1861, 1986.
15. Kurcok, P., Penczek, J., Franek, J., and Kedlinski, Z., Macromolecules, 25, 2285,
1992.
16. Nijenhuis, A.J., Du, Y.J., Van Aert, H.A.M., and Bastiaansen, C., Macromolecules,
28, 2124, 1995; Kricheldorf, H.R., Kreiser-Saunders, I., and Boettcher, C.,
Polymer, 36, 1253, 1995.
17. Sinclair, R.G., U.S. Patent 4,045,418, 1977; Sinclair, R.G., U.S. Patent 4,057,537,
1977.
18. Feijen, J., J. Appl. Polym. Sci., 62, 1295, 1996.
19. Ford, T.M., U.S. Patent 5,310,599, 1994.
20. Jacobsen, S., Fritz, H.G., Degee, Ph., Dubois, Ph., and Jerome, R., Polymer, 41,
3395, 2000.
21. Schmack, G., Jacobsen, S., Fritz, H.G., Vogel, R., Tandler, B., and Beyreuther, R.,
J. Biotechnol., 86, 15, 2001.
22. Jacobsen, S., Fritz, H. G., Degee, Ph., Dubois, Ph., and Jerome, R., J. Polym. Sci.,
37, 2413, 1999.
23. Jacobsen, S., Fritz, H.G., Degee, Ph., Dubois, Ph., and Jerome, R., Ind. Crops Prod.,
11, 265, 2000.
1741_C16.qxd
2/11/2005
9:58 AM
Polylactic Acid Technology
Page 571
571
24. Jacobsen, S., Fritz, H.G., Degee, Ph., Dubois, Ph., and Jerome, R., Macromol.
Symp., 144, 289, 1999.
25. Jacobsen, S., Fritz, H.G., Degee, Ph., Dubois, Ph., and Jerome, R., Macromol.
Symp., 153, 261, 2000.
26. Fukushima, T., Intern. Polym. Process., XV, 4,2000.
27. Miyoshi, R., Intern. Polym. Process., XI, 4, 1996.
28. Miyoshi, R., et al., U.S. Patent 5,574,129, 1996.
29. Narayan, R., et al., U.S. Patent 5,906,783, 1999.
30. Reichert, D., et al., U.S. Patent 5,378,801, 1995.
31. Jansson, K., Koskinen, J., and Selin, J.-F., U.S. Patent 6,214,967, 2001.
32. Witzke, D.R., Introduction to Properties, Engineering, and Prospects of
Polylactide Polymer, Ph.D. thesis, Chemical Engineering, Michigan State
University, 1997.
33. Garlotta, D., A literature review of poly(lactic acid), J. Polym. Environ., 9, 63,
2002.
34. Tsuji, H. and Ikada, Y., Physical properties of polylactides, Curr. Trends Polym.
Sci., 4, 27, 1999.
35. Tsuji, H., Polylactides, in Biopolymers 4 Polyesters III Applications and Commercial
Products, Wiley –VCH, Weinheim, Germany, 2002, chap. 5.
36. Hartmann, M.H., High molecular weight polylactic acid polymers, in
Biopolymers from Renewable Resources, Springer, Berlin, pp. 367–411.
37. Thakur, K.A.M., Kean, R.T., Hall, E.S., Doscotch, M.A., and Munson, E.J., A
quantitative method for determination of lactide composition in poly(lactide)
using 1H NMR, Anal. Chem., 69, 4303, 1997.
38. Thakur, K.M., Kean, R.T., Hall, E.S., Kolstad, J.J., and Munson, E.J.,
Stereochemical aspects of lactide stereo-copolymerization investigated by 1H
NMR: a case of changing stereospecificity, Macromolecules, 31, 1487, 1998.
39. Spassky, N., Wisniewski, M., Pluta, C., and Le Borgne, A., Highly stereoelective
polymerization of rac-(D,L)-lactide with a chiral Schiff’s base/aluminum alkoxide initiator, Macromol. Chem. Phys., 197, 2627, 1996.
40. Ovitt, T.M. and Coates, G.W., Stereochemistry of lactide polymerization with
chiral catalysts: new opportunities for stereocontrol using polymer exchange
mechanisms, J. Am. Chem. Soc., 124, 1316, 2002.
41. Ovitt, T.M. and Coates, G.W., Stereoselective ring-opening polymerization of
rac-lactide with a single-site, racemic aluminum alkoxide catalyst: synthesis of
stereoblock poly(lactic acid), J. Polym. Sci., Part A: Polym. Chem., 38, 4686, 2000.
42. Yui, N., Dijkstra, P., and Feijen, J., Stereo block copolymers of L- and D-lactides,
Makromol. Chem., 191, 487–488, 1990.
43. Ikada, Y., Jamshidi, K., Tsuji, H., and Hyon, S.H., Stereocomplex formation
between enantiomeric poly(lactides), Macromolecules 20, 904, 1987.
44. Fitz, B.D., Jamiolkowski, D.D., and Andjelic, S., Tg Depression in poly(L(-)-lactide) crystallized under partially constrained conditions, Macromolecules, 35,
5869, 2002.
45. Tsuji, H. and Ikada, Y., Stereocomplex formation between enantiomeric
poly(lactic acid)s. XI. Mechanical properties and morphology, Polymer, 40, 6699,
1999.
1741_C16.qxd
572
2/11/2005
9:58 AM
Page 572
Natural Fibers, Biopolymers, and Biocomposites
46. Urayama, H., Kanamori, T., and Kimura, Y., Properties and biodegradability of
polymer blends of poly(L-lactides) with different optical purity of the lactate
units, Macromole. Mater. Eng. 287, 116, 2002.
47. Janzen, J., Dorgan, J.R., Knauss, D.M., Hait, S.B., and Limoges, B.R.,
Fundamental solution and single-chain properties of polylactides,
Macromolecules, 2003.
48. Dorgan, J.R., Lehermeier, H., and Mang, M., Thermal and rheological properties
of commercial-grade poly(lactic acid)s, J. Polym. Environ., 8, 1, 2000.
49. Janzen, J. and Dorgan, J.R., A novel approach to modeling viscoelastic properties of thermoplastic polymer melts, with applications to polylactides, Annual
Technical Conf.– Soc. Plastics Engineers, 60, Vol. 1, 2002, p. 929.
50. Palade, L., Lehermeier, H.J., and Dorgan, J.R., Melt rheology of high L- content
poly(lactic acid), Macromolecules, 34, 1384, 2001.
51. Cooper-White, J.J. and MacKay, Michael, E., Rheological properties of poly(lactides). Effect of molecular weight and temperature on the viscoelasticity of
Poly(L-lactic acid), J. Polym. Sci., Part B: Polym. Phy., 37, 1803, 1999.
52. Ramkumar, D.H.S. and Bhattacharya, M., Steady shear and dynamic properties
of biodegradable polyesters, Polym. Eng. Sci., 38, 1426, 1998.
53. Feng, Q. and Hanna, M.A., Rheological properties of amorphous and semicrystalline polylactic acid polymers, Ind. Crops Prod., 10, 47, 1999.
54. Cicero, J.A., Dorgan, J.R., Dec, S.F., and Knauss, D.M., Phosphite stabilization
effects on two-step melt-spun fibers of polylactide, Polym Degradation Stability,
78, 95, 2002.
55. Gruber, P.R., Kolstad, J.J., Hall, E.S., Eichen-Conn, R.S., and Ryan, C.M., Meltstable lactide polymer compositions and their manufacture, PCT Int. Appl., WO
9407949, 1994.
56. Kang, S., Zhang, G., Aou, K., Hsu, S.L., Stidham, H.D., and Yang, X., An analysis
of poly(lactic acid) with varying regio regularity, J. Chem. Phy., 118, 3430, 2003.
57. Schindler, A., and Harper, D., Polylactide. II., Viscosity-molecular weight relationships and unperturbed chain dimensions., J. Polym. Sci., Polym. Chem. ed.,
17, 2593, 1979.
58. Schindler, A., Hibionada, Y.M., and Pitt, C.G., Aliphatic polyesters. III.
Molecular weight and molecular weight distribution in alcohol-initiated polymerizations of ε-caprolactone, J. Polym. Sci., Polym. Chem., 20, 319, 1982.
59. Dorgan, J.R., Williams, J.S., and Lewis, D.N., Melt rheology of poly(lactic acid):
Entanglement and chain architecture effects, J. Rheol., 43, 1141, 1999.
60. Zhao, Y., Cai, Q., Jiang, J., Shuai, X., Bei, J., Chen, C., and Xi, Fu., Synthesis and
thermal properties of novel star-shaped poly(L-lactide)s with starburst
PAMAM-OH dendrimer macroinitiator, Polymer, 43, 5819, 2002.
61. Drumright, R., Barker, H., Beaudoin, D., Berler, N., Broomall, C., Leibig, C.,
Hartmann, M., Henton, D., McCarthy, K., Olson, M., Randall, J., and Syverud,
K., Rheology modification of poly(lactic acid), ACS Regional Conference, Grand
Rapids, MI, 2001.
62. Tasaka, F., Ohya, Y., and Ouchi, T., One-pot synthesis of novel branched polylactide through the copolymerization of lactide with mevalonolactone,
Macromol. Rapid Commun., 22, 820, 2002.
1741_C16.qxd
2/11/2005
9:58 AM
Polylactic Acid Technology
Page 573
573
63. Drumright, R.E., Hartmann, M., and Wolf, R., Copolymers of monocyclic esters
and carbonates and methods for making same, Int. Appl., WO 02/100921A1,
Dec. 19, 2002.
64. Gruber, P.R., Kolstad, J.J., Witzke, D.R., Hartmann, M.H., and Brosch, A.L.,
Viscosity-Modified Lactide Polymer Composition and Process for Their
Manufacture, U.S. Patent 5,594,095, 1997.
65. Gruber, P.R., Kolstad, J.J., Ryan, C.M., Hall, E.S., and Eichen-Conn, R.S., Paper
having a melt-stable lactide polymer coating and process for manufacture, U.S.
Patent 5,338,822, 1995.
66. Lehermeier, H.J. and Dorgan, J.R., Melt rheology of poly(lactic acid): consequences of blending chain architectures., Polym. Eng. Sci., 41, 2172, 2001.
67. Huda, M. S., Yasui, M., Mohri, N., Fujimura, T., and Kimura, Y., Dynamic
mechanical properties of solution-cast poly(L-lactide) films, Mater. Sci. Eng., A:
Structural Mater: Prop., Microstruct. Process., A333, 98, 2002.
68. Abe, H., Kikkawa, Y., Inoue, Y., and Doi, Y., Morphological and kinetic analyses
of regime transition for poly[(S)-lactide] crystal growth, Biomacromolecules, 2,
1007, 2001.
69. Baratian, S., Hall, E.S., Lin, J.S., Xu, R., and Runt, J., Crystallization and solidstate structure of random polylactide copolymers: poly(L-lactide-co-D-lactide)s, Macromolecules, 34, 4857, 2001.
70. Zell, M.T., Padden, B.E., Paterick, A.J., Hillmyer, M.A., Kean, R.T., Thakur,
K.A.M., and Munson, E.J., Direct observation of stereodefect sites in semicrystalline poly(lactide) using 13C solid-state NMR, J. Am. Chem. Soc., 120, 12672,
1998.
71. Pyda, M., Buzin, A., Wunderlich, B., and Bopp, R.C., Morphology of poly(lactic
acid) by AFM and calorimetry, Proceedings of the 30th NATAS Annual Conference
on Thermal Analysis and Applications, 463, 2002.
72. Hoffman, D.J., Davis, G.T., and Lauritzen, J.I., Treatise on Solid State Chemistry,
Vol. 3: Crystalline and Noncrystaline Solids, Plenum Press, New York, 1976, chap.
7.
73. Tsuji, H. and Ikada, Y., Properties and morphologies of poly(L-lactide): 1.
Annealing condition effects on properties and morphologies of poly(L-lactide),
Polymer, 36, 2709, 1995.
74. Fischer, E.W., Sterzel, H.J., and Wegner, G., Investigation of the structure of
solution grown crystals of lactide copolymers by means of chemicals reactions.,
Kolloid Ze. Ze. fuer Polym., 251, 980, 1973.
75. Kolstad, J.J., Crystallization kinetics of poly(L-lactide-co-meso-lactide, J. Appl.
Polym. Sci., 62, 1079, 1996.
76. Pluta, M. and Galeski, A., Crystalline and supermolecular structure of polylactide in relation to the crystallization method, J. Appl. Polym. Sci., 86, 1386, 2002.
77. Schmidt, S.C. and Hillmyer, M.A., Polylactide stereocomplex crystallites as
nucleating agents for isotactic polylactide, J. Polym. Sci., Part B: Polym. Phy., 39,
300, 2001.
78. Liao, K., Quan, D., Gao, J., Luo, B., and Lu, Z., Crystallization and Miscibility of
Poly(L-lactide)/Poly(DL-lactide) Blends, Institute of Polymer Science,
Zhongshan University, Canton, People Republic of China.
1741_C16.qxd
574
2/11/2005
9:58 AM
Page 574
Natural Fibers, Biopolymers, and Biocomposites
79. Spinu, M., Jackson, C., Keating, M.Y., and Gardner, K.H., Material design in
poly(lactic acid) systems: block copolymers, star homo- and copolymers, and
stereocomplexes, J. Macromol. Sci., Pure. Appl. Chem., A33, 1497, 1996.
80. Tsuji, H. and Ikada, Y., Crystallization from the melt of poly(lactide)s with different optical purities and their blends, Macromol. Chem. Phy., 197, 3483, 1996.
81. Tsuji, H. and Ikada, Y., Stereocomplex formation between enantiomeric
poly(lactic acid)s. XI. Mechanical properties and morphology of solution-cast
films, Polymer, 40, 6699, 1999.
82. Loomis, G.L. and Murdoch, J.R., Polylactide stereocomplexes, Polym. Prepr. 31,
55, 1990.
83. Okihara, T., Kawaguchi, A., Tsuji, H., Hyon, S.H., Ikada, Y., and Katayama, K.,
Lattice disorders in the stereocomplex of poly(L-lactide) and poly(D-lactide),
Bull. Inst. Chem. Res., Kyoto Univ., 66, 271, 1988.
84. Tsuji, H., Horii, F., Hyon, S.H., and Ikada, Y., Stereocomplex formation between
enantiomeric poly(lactic acid)s. 2. Stereocomplex formation in concentrated
solutions, Macromolecules, 24, 2719, 1991.
85. Okihara, T., Tsuji, M., Kawaguchi, A., Katayama, K., Tsuji, H., Hyon, S.H., and
Ikada, Y., Crystal structure of stereocomplex of poly(L-lactide) and poly(D-lactide), J. Macromol. Sci., Phy., 30, 119, 1991.
86. Tsuji, H., Hyon, S.H., and Ikada, Y., Stereocomplex formation between enantiomeric poly(lactic acid)s. 3. Calorimetric studies on blend films cast from
dilute solution, Macromolecules, 24, 5651, 1991.
87. Tsuji, H., Hyon, S.H., and Ikada, Y., Stereocomplex formation between enantiomeric poly(lactic acid)s. 4. Differential scanning calorimetric studies on precipitates from mixed solutions of poly(D-lactic acid) and poly(L-lactic acid),
Macromolecules, 24, 5657, 1991.
88. Tsuji, H., Hyon, S.H., and Ikada, Y., Stereocomplex formation between enantiomeric poly(lactic acids). 5. Calorimetric and morphological studies on the
stereocomplex formed in acetonitrile solution, Macromolecules, 25, 2940, 1992.
89. Tsuji, H., Horii, F., Nakagawa, M., Ikada, Y., Odani, H., and Kitamaru, R.,
Stereocomplex formation between enantiomeric poly(lactic acid)s. 7. Phase
structure of the stereocomplex crystallized from a dilute acetonitrile solution as
studied by high-resolution solid-state carbon-13 NMR spectroscopy,
Macromolecules, 25, 4114, 1992.
90. Tsuji, H. and Ikada, Y., Stereocomplex formation between enantiomeric
poly(lactic acid)s. 6. Binary blends from copolymers, Macromolecules, 25, 5719,
1992.
91. Tsuji, H. and Ikada, Y., Stereocomplex formation between enantiomeric
poly(lactic acids). 9. Stereocomplexation from the melt, Macromolecules, 26, 6918,
1993.
92. Tsuji, H., Ikada, Y., Hyon, S.H., Kimura, Y., and Kitao, T., Stereocomplex formation between enantiomeric poly(lactic acid). VIII. Complex fibers spun from
mixed solution of poly (D-lactic acid) and poly (L-lactic acid), J. Appl. Polym.
Sci., 51, 337, 1994.
93. Tsuji, H., In vitro hydrolysis of blends from enantiomeric poly(lactide)s part 1.
Well-stereocomplexed blend and non-blended films, Polymer, 41, 3621, 2000.
1741_C16.qxd
2/11/2005
9:58 AM
Polylactic Acid Technology
Page 575
575
94. Tsuji, H. and Suzuki, M. In vitro hydrolysis of blends from enantiomeric
poly(lactide)s. 2. Well-stereocomplexes fiber and film. Sen’i Gakkaishi, 57, 198,
2001.
95. Kang, S., Aou, K., Pekrul, R., and Hsu, S.L., Crystallization of poly(lactic acid)
stereocomplexes, 224th ACS National Meeting, Boston, MA, US, Abstr., Aug.
18–22, 2002.
96. Kokturk, G., Serhatkulu, T.F., Cakmak, M., and Piskin, E., Evolution of phase
behavior and orientation in uniaxially deformed polylactic acid films, Polym.
Eng. Sci., 42, 1619, 2002.
97. Puiggali, J., Ikada, Y., Tsuji, H., Cartier, L., Okihara, T., and Lotz, B., The frustrated structure of poly(L-lactide), Polymer, 41, 8921, 2000.
98. Lee, J.K., Lee, K.H., and Jin, B.S., Structure development and biodegradability
of uniaxially stretched poly(L-lactide), Eur. Polym. J., 37, 907, 2001.
99. Kister, G., Cassanas, G., Vert, M., Pauvert, B., and Terol, A., Vibrational analysis
of poly(L-lactic acid), J. Raman Spectrosc., 26, 307, 1995.
100. Smith, P.B., Leugers, A., Kang, S., Yang, X., and Hsu, S.L., Raman characterization of orientation in poly(lactic acid) films, Macromol. Symp. Polymerization
Processes Polym. Mater. II, 175, 81, 2001.
101. Smith, P.B., Leugers, A., Kang, S., Hsu, S.L., and Yang, X., An analysis of the correlation between structural anisotropy and dimensional stability for drawn
poly(lactic acid) films, J. Appl. Polym. Sci., 82, 2497, 2001.
102. Kang, S., Hsu, S.L., Stidham, H.D., Smith, P.B., Leugers, A.M., and Yang, X., A
Spectroscopic analysis of poly(lactic acid) structure, Macromolecules, 34, 4542,
2001.
103. Randall, J., McCarthy, K., Krueger, J., Smith, P., and Spruiell, J., Properties of
Polylactic Acid Optical Copolymers Achieved Through Stress Induced
Crystallization, 58th Annual Technical Conference - Society of Plastics
Engineers, Vol. 3, 3728, 2000.
104. Cicero, J.A., Dorgan, J.R., Garrett, J., Runt, J., and Lin, J.S., Effects of molecular
architecture on two-step, melt-spun poly(lactic acid) fibers, J. Appl. Polym. Sci.,
86, 2839, 2002.
105. Cicero, J.A., Dorgan, J.R., Janzen, J., Garrett, J., Runt, J., and Lin, J.S.,
Supramolecular morphology of two-step, melt-spun poly(lactic acid) fibers,
J. Appl. Polym. Sci., 86, 2828, 2002.
106. Keller, A., Barham, P., and Wills, H., J. Polym. Sci., 13, 197, 1975.
107. Lunt, J., Large-scale production, properties and commercial applications of
polylactic acid polymers, Polym. Degrad. Stability, 59, 145, 1998.
108. Tsuji, H., Polylactides, in Biopolymers, Wiley-VCH, 2002, chap. 5.
109. Pitt, C.G., Jeffcoat, A.R., and Zweidinger, R.A., Sustained drug delivery system.
I. The permeability of poly(caprolactone), poly (DL-lactic acid), and their
copolymers, J. Biomed. Mater. Res., 13, 497, 1979.
110. Siparsky, G., Voorhees, K., and Miao, F., J. Environ. Polym. Degrad., 6, 31, 1998.
111. Lee, S., Kim, S.H., Han, Y., and Kim, Y., Synthesis and degradation of end-groupfunctionalized polylactide, J. Polym. Sci., Part A: Polym. Chem., 39, 973, 2001.
112. Joziasse, C.A.P., Grijpma, D.W., Bergsma, J.E., Cordewener, F.W., Bos, R.R.M.,
and Pennings, A.J., The influence of morphology on the hydrolytic degradation
1741_C16.qxd
576
113.
114.
115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
126.
127.
128.
129.
2/11/2005
9:58 AM
Page 576
Natural Fibers, Biopolymers, and Biocomposites
of as-polymerized and hot-drawn poly(L-lactide), Colloid Polym. Sci., 276, 968,
1998.
Grizzi, I., Garreau, H., Li, S., and Vert, M., Hydrolytic degradation of devices
based on poly(DL-lactic acid) size-dependence, Biomaterials, 16, 305, 1995.
Nakamura, T., Hitomi, S., Watanabe, S., Shimizu, Y., Jamshidi, K., Hyon, S.H.,
and Ikada, Y., Bioabsorption of polylactides with different molecular properties,
J. Biomed. Mater. Res., 23, 1115, 1989.
Vert, M., Li, S., Garreau, H., Mauduit, J., Boustta, M., Schwach, G., Engel, R., and
Coudane, J., Complexity of the hydrolytic degradation of aliphatic polyesters,
Ange. Makromol. Chem., 247, 239, 1997.
Mauduit, J., Perouse, E., and Vert, M., Hydrolytic degradation of films prepared
from blends of high and low molecular weight poly(DL-lactic acids), J. Biomed.
Mater. Res., 30, 201, 1996.
Hyon, S., Jamshidi, K., and Ikada, Y., Effects of residual monomer on the degradation of DL-lactide polymer, Polym. Int., 46, 196, 1998.
Iwata, T. and Doi, Y., Alkaline hydrolysis of solution-grown poly(L-lactic acid)
single crystals, Sen’i Gakkaishi, 57, 172, 2001.
Tsuji, H. and Nakahara, K., Poly(l-lactide). IX. Hydrolysis in acid media, J. Appl.
Polym. Sci., 86, 186, 2002.
Makino, K. et al., Preparation and in vitro degradation properties of polylactide
microcapsules, Chem. Pharm. Bull., 33, 1195, 1985.
Yuan, X., Mak, A., and Yao, K., In vitro degradation of poly(L-lactic acid) fibers
in phosphate buffered saline, J. Appl. Polym. Sci., 85, 936, 2002.
Tsuji, H. and Suzuyoshi, K., Environmental degradation of biodegradable polyesters 2. Poly(epsiln.-caprolactone), poly[(R)-3-hydroxybutyrate], and poly(L-lactide) films in natural dynamic seawater, Polym. Degrad. Stability, 75, 357, 2002.
Tsuji, H., Nakahara, K., and Ikarashi, K., Poly(L-lactide), 8a. High-temperature
hydrolysis of poly(L-lactide) films with different crystallinities and crystalline
thicknesses in phosphate-buffered solution, Macromol. Mater. Eng., 286, 398,
2001.
Iannace, S., Maffezzoli, A., Leo, G., and Nicolais, L., Influence of crystal and
amorphous phase morphology on hydrolytic degradation of PLLA subjected to
different processing conditions, Polymer, 42, 3799, 2001.
Amass, W., Amass, A., and Tighe, B., A review of biodegradable polymers: uses,
current developments in the synthesis and characterization of biodegradable
polyesters, blends of biodegradable polymers and recent advances in biodegradation studies, Polym. Int., 47, 89, 1998.
Shih, C., A graphical method for the determination of the mode of hydrolysis of
biodegradable polymers, Pharm. Res., 12, 2036, 1995.
Zhang, X., Wyss, U., Pichora, D., and Goosen, M., Investigation of poly(lactic
acid) degradation, J. Bioact. Compat. Polym., 9, 80, 1994.
Li, S.M. and Vert, M., Degradable Polymers: Principles and Applications, Chapman
& Hall, London, 1995.
Sharp, J. S., Forrest, J.A., and Jones, R.A.L., Swelling of poly(DL-lactide) and
polylactide-co-glycolide in humid environments. Macromolecules, 34, 8752,
2001.
1741_C16.qxd
2/11/2005
9:58 AM
Polylactic Acid Technology
Page 577
577
130. McCarthy, K., Cargill Dow LLC product literature, Minnetonka, Minnesota,
1998.
131. Siparsky, G.L., Voorhees, K.J., Dorgan, J.R., and Schilling, K., Water transport in
polylactic acid (PLA), PLA/polycaprolactone copolymers, and PLA/polyethylene glycol blends, J. Environ. Polym. Degrad., 5, 125, 1997.
132. Li, S. and McCarthy, S., Further investigations on the hydrolytic degradation of
poly (DL-lactide). Biomaterials, 20, 35, 1999.
133. Burg, K.J.L. and Shalaby, S.W., Water fugacity in absorbing polymers, J. Biomed.
Mater. Res., 38, 337, 1997.
134. Schmitt, E.A., Flanagan, D.R., and Linhardt, R.J., Importance of distinct water
environments in the hydrolysis of poly(DL-lactide-co-glycolide),
Macromolecules, 27, 743, 1994.
135. Renstad, R., Karlsson, S., Sandgren, A., and Albertsson, A., Influence of processing additives on the degradation of melt-pressed films of poly(ε-caprolactone) and poly(lactic acid), J. Environ. Polym. Degrad., 6, 209, 1998.
136. Li, S. and Vert, M., Hydrolytic degradation of coral/poly(DL-lactic acid) bioresorbable material. J. Biomater. Sci., Polymer ed., 7, 817, 1996.
137. Fischer, E.W., Structure of partially crystalline polymer systems, Prog. Colloid
Polym. Sci., 57, 149, 1975.
138. Nijenhuis, A.J., Grijpma, D.W., and Pennings, A.J., Highly crystalline as-polymerized poly(L-lactide), Polym. Bull., 26, 71, 1991.
139. Tsuji, H. and Miyauchi, S., Poly(L-lactide): VI effects of crystallinity on enzymatic hydrolysis of poly(L-lactide) without free amorphous region, Polym.
Degrad. Stability, 71, 415, 2000.
140. Tsuji, H. and Miyauchi, S., Poly(L-lactide): 7. enzymatic hydrolysis of free and
restricted amorphous regions in poly(L-lactide) films with different crystallinities and a fixed crystalline thickness, Polymer, 42, 4463, 2001.
141. United States Environmental Protection Agency Presidential Green Chemistry
Challenge, Alternate Solvents/Reaction Conditions Award, June 2002.
1741_C16.qxd
2/11/2005
9:58 AM
Page 578