See

discussions, stats, and author profiles for this publication at:

https://www.researchgate.net/publication/234076748

The chemomechanical properties of

microbial polyhydroxyalkanoates

Conference Paper in Progress in Polymer Science · January 2012

Impact Factor: 26.93 · DOI: 10.1016/j.progpolymsci.2012.06.003

CITATIONS READS

89 180

5 authors, including:

Peter J. Halley Steven Pratt

University of Queensland University of Queensland
272 PUBLICATIONS 3,651 CITATIONS 69 PUBLICATIONS 1,181 CITATIONS

SEE PROFILE SEE PROFILE

Alan G Werker Paul Lant

AnoxKaldnes AB University of Queensland
68 PUBLICATIONS 1,021 CITATIONS 135 PUBLICATIONS 3,758 CITATIONS

SEE PROFILE SEE PROFILE

Available from: Bronwyn Laycock

Retrieved on: 04 May 2016
 Progress in Polymer Science 39 (2014) 397–442

Contents lists available at ScienceDirect

Progress in Polymer Science

journal homepage: www.elsevier.com/locate/ppolysci

Review

The chemomechanical properties of microbial

polyhydroxyalkanoates
Bronwyn Laycock a,b,∗ , Peter Halley a,b,1 , Steven Pratt a,2 ,
Alan Werker c,3 , Paul Lant a,4
a
School of Chemical Engineering, The University of Queensland, St Lucia, Qld 4072, Australia
b
Australian Institute for Bioengineering and Nanotechnology, The University of Queensland, Brisbane, Qld 4072, Australia
c
AnoxKaldnes AB, Klosterängsvägen 11A, SE-226 47 Lund, Sweden

a r t i c l e i n f o a b s t r a c t

Article history: Microbially produced polyhydroxyalkanoates (PHAs) are fully biodegradable biopolyesters
Received 22 May 2013 that have attracted much attention recently as alternative polymeric materials that can be
Accepted 3 June 2013
produced from biorenewable and biowaste resources. The properties of these biological
Available online 21 August 2013
polymers are affected by the same fundamental principles as those of fossil-fuel derived
polyolefins, with a broad range of compositions available based on the incorporation of
Keywords:
different monomers into the PHA polymer structure, and with this broad range tailoring
Biopolymer
Polyhydroxyalkanoates subsequent properties. This review comprehensively covers current understanding with
Microbial polyesters respect to PHA biosynthesis and crystallinity, and the effect of composition, microstructure
Poly((R)-3-hydroxybutyrate) and supramacromolecular structures on chemomechanical properties. While polymer com-
Poly((R)-3-hydroxybutyrate-co-3- position and microstructure are shown to affect these properties, the review also finds that
hydroxyvalerate) a key driver for determining polymer performance properties is compositional distribution.
Mechanical properties From this review it follows that PHA–PHA blend compositions are industrially important,
and the performance properties of such blends are discussed. A particular need is identified
for further research into the effect of chemical compositional distribution on macromolec-
ular structure and end-use properties, advanced modeling of the PHA accumulation process
and chain growth kinetics for better process control.
© 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Bacterial polyhydroxyalkanoates (PHAs) are a unique

family of polymers that act as a carbon/energy store
for more than 300 species of Gram-positive and Gram-
DOI of the original article: negative bacteria as well as a wide range of archaea [1].
http://dx.doi.org/10.1016/j.progpolymsci.2012.06.003.
∗ Corresponding author at: School of Chemical Engineering, The Uni-
Synthesized intracellularly as insoluble cytoplasmic inclu-
versity of Queensland, St Lucia, Qld 4072, Australia. Tel.: +61 7 3346 3188;
sions in the presence of excess carbon when other essential
fax: +61 7 3346 3973. nutrients such as oxygen, phosphorous or nitrogen are
E-mail addresses: b.laycock@uq.edu.au (B. Laycock), limited, these polymeric materials are able to be stored at
p.halley@uq.edu.au (P. Halley), s.pratt@uq.edu.au (S. Pratt), high concentrations within the cell since they do not sub-
Alan.Werker@anoxkaldnes.com (A. Werker), paul.lant@uq.edu.au
stantially alter its osmotic state [2]. The resulting polymers
(P. Lant).
1
Fax: +61 7 3346 3973.
are piezoelectric and perfectly isotactic/optically active
2
Fax: +61 7 3346 7843. (having only the (R)-configuration). They are hydrophobic,
3
Fax: +46 46 133201. water-insoluble, inert and indefinitely stable in air, and
4
Fax: +61 7 3365 4728. are also thermoplastic and/or elastomeric, non-toxic and

0079-6700/$ – see front matter © 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.progpolymsci.2013.06.008
398 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442

Nomenclature
kDa kilodalton
2D two-dimensional lc lamellar core distance
3D three-dimensional Lp long period distance
13 C carbon-13 Mn number-average molecular weight
3HB 3-hydroxybutyrate Mw weight-average molecular weight
3HD 3-hydroxydecanoate mcl medium chain length
3HDD 3-hydroxydodecanoate MDa megadalton
3HHx 3-hydroxyhexanoate MPa megapascals
3H2MB 3-hydroxy-2-methylbutyrate ␮m micrometer
3H2MV 3-hydroxy-2-methylvalerate nA Avrami index
3H4MV 3-hydroxy-4-methylvalerate NADH nicotinamide adenine dinucleotide
3HO 3-hydroxyoctanoate (reduced form)
3HP 3-hydroxypropionate NADPH nicotinamide adenine dinucleotide phos-
3HPE 3-hydroxy-4-pentenoate phate (reduced form)
3HTD 3-hydroxytetradecanoate nm nanometer
3HV 3-hydroxyvalerate NMR nuclear magnetic resonance
4HB 4-hydroxybutyrate P(3HB) poly(3-hydroxybutyrate)
ACP acyl carrier protein P(3HB-co-3HV) poly(3-hydroxybutyrate-co-3-
Ae aerobic hydroxyvalerate)
ADF aerodynamic feeding P(3HB-co-4HB) poly(3-hydroxybutyrate-co-4-
AFM atomic force microscopy hydroxybutyrate)
An anaerobic P(3HB-co-3HHx) poly(3-hydroxybutyrate-co-3-
BOD biochemical oxygen demand hydroxyhexanoate)
CCD chemical compositional distribution PAOs polyphosphate accumulating organisms
CH2 methylene group PDI polydispersity index
CoA coenzyme A PEG polyethylene glycol
COD chemical oxygen demand PHA polyhydroxyalkanoate
Da Dalton PHO polyhydroxyoctanoate
DMTA dynamic mechanical thermal analysis Pn number average degree of polymerization
DO dissolved oxygen R R enantiomeric form
DOC dissolved organic carbon concentration RNA ribonucleic acid
DSC differential scanning calorimetry SBR sequencing batch reactor
DP degree of polymerization scl short chain length
Ea activation energy SDS-PAGE sodium dodecyl sulfate polyacrylamide
EBPR enhanced biological phosphorus removal gel electrophoresis
ESI-MS electrospray ionization multistage mass t time
spectrometry Tc crystallization temperature
FBB fraction of the BB diad sequence in a P(3HB- Tcc cold crystallization temperature
co-3HV) copolymer Tg glass transition temperature
FBV fraction of the BV diad sequence in a P(3HB- Tm melting temperature
co-3HV) copolymer Tm0 equilibrium melting temperature
FVB fraction of the VB diad sequence in a P(3HB- TCA tricarboxylic acid cycle
co-3HV) copolymer TEM transmission electron microscopy
FVV fraction of the VV diad sequence in a P(3HB- UHMW ultra high molecular weight
co-3HV) copolymer UV ultraviolet
FAB-MS fast atom bombardment mass spectroscopy VFAs volatile fatty acids
FT-IR Fourier-transform infrared WAXD wide angle X-ray diffraction
g gram WAXS wide angle X-ray scattering
G growth rate of spherulites WWTP wastewater treatment plant
GAOs glycogen accumulating organisms Xt percent crystallinity
GPa gigopascals XRD X-ray diffraction
GPC gel permeation chromatography
Hf heat of fusion of the melting peak
Hf 100% heat of fusion of a 100% crystalline material have very high purity within the cell [3–8]. PHA has a
HB hydroxybutyrate much better resistance to UV degradation than polypro-
HDPE high density polyethylene pylene but is less solvent resistant. Most importantly,
HV hydroxyvalerate these biopolymers are completely biodegradable. To
kA crystallization rate constant date more than 150 different monomer units have been
incorporated into biological PHA [3–8], and the polymer
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
properties available within this family are as a result very
broad. In general, PHAs can be divided into two main
groups, these being the short-chain-length PHAs (scl-
PHAs) that contain monomer units with 3–5 carbon atoms,
and the medium-chain-length PHAs (mcl-PHAs) that
contain monomer units of 6–18 carbon atoms. The most
common PHAs are poly(3-hydroxybutyrate) (P3HB) and
poly(3-hydroxybutyrate-co-3-hydroxyvalerate) (P(3HB-
co-3HV)). These materials have mechanical properties that
are comparable to those of polypropylene and polyethyl-
ene, although they have much lower elongation to break
and are more brittle.
There are a number of reviews of PHA polymers and
their crystallinity [9–26]. However, the effect of micro-
structure and composition, and in particular the influence
of compositional distribution/blending, on mechanical
properties has not been a focus. This is addressed in this
review. Special attention is paid to higher HV content
copolymers (>20 mol% 3HV), the properties of which have
been infrequently reported. The fundamentals of PHA bio-
chemistry and crystallinity are given as a background,
with the scope of the review being summarized in Fig. 1.
It is believed that a clear understanding of the relation-
ships between PHA crystallinity and polymer composition,
compositional distribution, microstructure and blend com-
position will help inform future polymer development
within this field.
2. Overview of PHA production
2.1. PHA production in pure cultures
The French bacteriologist Lemoigne first isolated
and characterized poly-3-hydroxybutyrate (P(3HB)) from
bacteria between 1923 and 1927 [27–33], and showed
that this extract could be cast into a transparent film.
Although it was some time before this discovery was
turned into a practical outcome, PHA production using
pure cultures still has a long history. Stanier and Wilkin-
son and their co-workers were responsible for some of
the initial fundamental research into the mechanisms of
PHA biosynthesis, beginning in 1957 [34,35]. This was fol-
lowed in 1959 by the first attempted commercialization,
when W.R. Grace and Company patented a P(3HB) pro-
duction process using bacteria [36], although production
inefficiencies, poor thermal stabilities and a lack of avail-
able extraction technologies limited this application. In
1970, Imperial Chemical Industries Ltd. commercialized
the production of P(3HB-co-3HV) under the trade name of
BiopolTM [37], with the technology since being sold to Mon-
santo and then to Metabolix. Since that time, there have
been a range of new technologies developed, and recent
focus within pure culture production has been on the syn-
thesis of alternative copolymers such as P(3HB-co-3HHx)
(commercialized as NodaxTM ) [38], a wide range of func-
tionalized PHAs (incorporating monomer units with novel
and active functionalities on the side chains) and also on
the use of metabolic engineering to reengineer the cen-
tral metabolism of PHA producers for more efficient PHA
production. In its simplest form, PHA production using
pure cultures adopts a two-stage batch production process
399
[22], with an inoculum of bacteria being introduced into
a sterile solution of trace metal nutrients and a suitable
carbon source and nutrients in the first (growth) stage. In
the second stage, an essential nutrient (such as N, P or O2 )
is deliberately limited and PHA accumulation takes place.
The properties of the final polymer depends on the mix of
carbon feedstocks fed during accumulation, the metabolic
pathways that the bacteria use for the following conver-
sion into precursors, and the substrate specificities of the
enzymes involved [39].
2.2. PHA production in mixed cultures
PHA production based on open mixed cultures has
been proposed as a means of lowering production costs
[40,41]. No reactor sterilization is necessary and the cul-
ture is able to adapt to various complex (cheap) waste
feedstocks [40–46]. Gurieff and Lant [47] undertook a life
cycle assessment and financial analysis of mixed culture
PHA production using an industrial wastewater as the feed-
stock, and found that PHA production was preferable to
biogas production and was financially attractive in com-
parison to pure culture PHA production, while both PHA
production processes had similar environmental impacts
that were significantly lower than HDPE production.
The synthesis of PHA in mixed cultures was first
observed in 1974 in wastewater treatment plants designed
for biological phosphorus removal (EBPR) [48], when
Wallen and Rohwedder reported PHA heteropolymers in
the chloroform extracts of the activated sewage sludge.
Most mixed culture production relies on ecological selec-
tion pressures that favor organisms with elevated PHA
storage capacity, thus engineering the microbial consor-
tium [45,49,50]. One such method is to enrich a culture
in PHA-accumulating organisms by taking the culture
through a number of aerobic feast/famine cycles (known as
the aerobic dynamic feeding, or ADF, process) [51,52]. ADF
relies on the removal of a portion of the consortium dur-
ing each cycle – the organisms that are able to store carbon
during feast condition and then use that stored carbon to
grow during famine conditions are selectively enriched. A
modified approach is to use alternating aerobic/anaerobic
conditions; in this case it is the polyphosphate and glyco-
gen accumulating organisms (PAOs and GAOs) that are
selectively enriched and therefore responsible for PHA
accumulation [53,54].
Volatile fatty acids (VFAs) can be readily stored as PHA
by mixed cultures, so most studies on PHA production
by mixed cultures have been carried out using organic
acids (such as acetate, propionate, butyrate, and valerate)
[55–58] as feedstocks. Fatty acids are energetically advan-
tageous substrates from a metabolic perspective since their
complete ␤-oxidation generates more equivalent chemical
energy than the complete oxidation of a molar equiva-
lent of glucose [59]. A major advantage of mixed culture
PHA production is the opportunity to use real fermented
wastes as feedstock as opposed to synthetic VFAs, since
substrate costs are known to be a critical factor in deter-
mining the economics of PHA production. The use of low
cost agricultural or industrial waste feedstocks such as
fermented sugar cane molasses [42,53,60–62], fermented
400 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442

Fig. 1. Overview of PHA synthesis: (a) schematic depiction of chain polymerisation catalyzed by enzymes. (b) Schematic depiction of a PHA granule with
granule associated proteins. Reproduced with permission, from Rehm (2003), Biochemical Journal, vol. 376, page 27 © the Biochemical Society [1]. (c)
Schematic depiction of the ␣- and ␤-forms of the PHB polymer chain. (d) Schematic depiction of semi-crystalline polymer structure. (e) AFM image of
PHBV film, courtesy of JPK Instruments AG. (f) Final plastic products.

paper mill effluents [43], saponified sunflower oil [63], fruit
[64] and tomato [65] cannery effluents, fermented munici-
pal sludge [66,67] fermented food industrial and domestic
wastewaters [68,69], bio-oil (a pyrolysis by-product) [70]
and fermented olive oil mill effluents [71,72] have all so far
resulted in the successful accumulation of PHA, and there
is a growing knowledge base in the use of other feedstocks
such as alcohols and glycerol [11].
A broad range of PHA compositions can be eas-
ily produced using mixed cultures based on differ-
ing feedstocks, with other monomer units such as
3-hydroxyvalerate (3HV), 3-hydroxy-2-methylbutyrate
(3H2MB), 3-hydroxy-2-methylvalerate (3H2MV), and 3-
hydroxyhexanoate (3-HHx) being common components
in P(3HB)-based copolymers [58,62,74–76]. For example,
P(3HB-co-3HV) copolymers with mol% HV ranging from
17 to 85% have been recorded based on the propionate
fraction in acetate/propionate feed [58,75] (see Appendix
A). Pure cultures by contrast often need large amounts of
co-substrates to produce polymers with relatively small
fractions of monomers other than P(3HB) (see for exam-
ple [76] where P(3HB-co-3HV) with only 2–8 mol% HV was
produced from mixtures containing equal amounts of veg-
etable oils and propionic acid (used to produce HV)). This
may be due to the fact that mixed cultures contain a diver-
sity of organisms that are likely to employ a range of PHA
production pathways. Cellular PHA contents and produc-
tion rates comparable to or superior to those of pure culture
have now been achieved by Johnson et al. [77,78] using
mixed cultures, with this team producing P(3HB) at 89 wt%
of the dried biomass in less than 8 h.
Further research is required to understand the effect of
key controlling factors such as feeding strategy (continu-
ous or pulse) [79], pH control [80] and microstructure [81]
in mixed cultures, as they may have quite different effects
under these production conditions compared to pure cul-
ture PHA production. Given the multiple organisms (and
possibly also PHA production pathways) present in a mixed
culture environment, one of the key questions will be the
compositional distribution of the polymer produced and
effect of these blend compositions on chemomechanical
properties.
2.3. Other production techniques
Plant-based (photosynthesis fuelled) production sys-
tems for PHA have been under intensive investigation
over recent years [82,83]. Poirier first demonstrated
this approach in 1992 [84] using transgenic Arabidopsis
expressing acetoacetyl-CoA reductase and PHB synthase
from Ralstonia eutropha. (There have been a series of name
changes for many of the organisms used for PHA accumula-
tion. For example, Ralstonia eutropha (the most extensively
studied bacterium) was originally named Hydrogenomonas
eutrophus, then renamed Alcaligenes eutropha, and has
since been renamed.) The commercial development of
plant-based PHA production processes is now underway,
although there is still much research to be done. For exam-
ple, Metabolix has been developing technologies in a range
of transgenic plants, such as tobacco (Nicotiana tabacum)
[85] and switchgrass [86]. Preliminary investigations are
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
also underway into the production of PHA using microalgae
[87].
Finally, PHAs can be synthesized from the relevant sub-
stituted propiolactones using, for example, aluminum or
zinc alkyl catalysts with water as a cocatalyst (see [88,89]
for example). However, this process is not cost-competitive
for large-scale production and involves the use of toxic
reagents.
2.4. Conclusions
There are many approaches to the production of PHA
with potential variation in composition and compositional
distribution depending on the microbial diversity of the
culture used, etc. Regardless of the production strategy
being adopted, it is of fundamental importance to under-
stand the underlying biological processes and chemistries
that affect the chemomechanical properties of the final
polymer product. These are reviewed in the following sec-
tions.
3. PHA fundamentals
3.1. Biosynthesis
The biology of PHA bioplastics is complex; PHA accu-
mulation is controlled by many genes that encode a range
of enzymes that are directly or indirectly involved in
PHA synthesis [5,10,13,23,25,26,90]. So far, biosynthesis
of PHA can be summarized in eight pathways [12]. The
first pathway (Pathway I) is well-studied, and is used
by the model organism Ralstonia eutropha; it involves
the three key enzymes ␤-ketothiolase, NADPH-dependent
acetoacetyl-CoA reductase, and PHA synthase, encoded by
genes phaA, phaB and phaC, respectively. These three genes
are found together on an operon whose expression is rela-
tively constant during PHA production [91,92]. The carbon
source is initially converted into coenzyme A thioesters
of (R)-hydroxyalkanoic acid. ␤-Ketothiolase can then cat-
alyze the condensation of two coenzyme A thioester
monomers (such as an acetyl-CoA and a propionyl-CoA
monomer). This is followed by an (R)-specific reduction to
give (R)-3-hydroxybutyryl-CoA (or (R)-3-hydroxyvaleryl-
CoA) (catalyzed by acetoacetyl-CoA reductase), which is
then converted by PHA synthase into PHA [91,93]. At
least 88 PHA synthases have been sequenced, with four
major classes being identified [94]. Class I uses fatty acids
with 3–5 carbon atoms; class II uses those with 6–14
carbons; and Classes III and IV synthesize short chain
length PHA. In the depolymerization reaction the accu-
mulated PHA is hydrolyzed into 3-hydroxybutyrate (3HB)
by depolymerase (encoded by phaZ), and can then be
converted back into acetoacetyl-coenzyme A. Pathway
II is associated with fatty acid uptake by microorgan-
isms and can be used for the synthesis of mcl-PHA.
Following fatty acid ␤-oxidation to give acyl-CoA, the pre-
cursor is then converted to 3-hydroxyacyl-CoA which can
then form PHA under synthase catalysis. Known enzymes
involved in this pathway include 3-ketoacyl-CoA reduc-
tase, epimerase and (R)-enoyl-CoA hydratase/enoyl-CoA
hydratase I. Two others are also believed to be involved:
401
acyl-CoA oxidase and enoyl-CoA hydratase. Pathway III
involves 3-hydroxyacyl-ACP-CoA transferase (encoded by
PhaG) and malonyl-CoA-ACP transacylase (FabD); sub-
strates are converted to 3-hydroxyacyl-ACP which can
then form 3-hydroxyacyl-CoA and thus PHA. Pathway IV
uses NADPH-dependent acetoacetyl-CoA reductase to oxi-
dize (S)-(+)-3-hydroxybutyryl-CoA. The other pathways
are used for the synthesis of alternative copolymers; for
example, pathways V and VII are used to synthesize P(4HB)
containing PHAs (and are used by Clostridium kluyveri and
A. hydrophila 4AK4 respectively).
As Kessler and Witholt state: “Regulation of PHA
metabolism can take place at different levels: (1) activa-
tion of pha gene expression due to specific environmental
signals, such as nutrient starvation; (2) activation of
the PHA synthetic enzymes by specific cell components
or metabolic intermediates; (3) inhibition of metabolic
enzymes of competing pathways and therefore enrichment
of required intermediates for PHA synthesis; or (4) a combi-
nation of these.” [23]. For example, during normal bacterial
growth, the ␤-ketothiolase in Pathway I is inhibited by
free coenzyme-A coming out of the Krebs (or TCA) cycle.
But when nutrients other than carbon are limited, acetyl-
CoA is restricted from entering into the Krebs cycle and
the excess acetyl-CoA is channeled into PHA biosynthesis
[95]. If growth is limited for other reasons, protein syn-
thesis stops, which leads to high concentrations of NADH
and NADPH. This inhibits citrate synthase and isocitrate
dehydrogenase, which again slows down the Krebs cycle,
directing acetyl-CoA towards PHA synthesis [96].
The synthesis of the PHA polymer chain takes place
within the cytoplasm of the bacterial cell, within inclusions
known as granules – the biochemistry of this synthetic pro-
cess and of these granules in turn influences the properties
of the final polymer product.
3.2. Granule structure and development
Granules in microorganisms were observed under the
microscope by Beijerinck as early as 1888 (reported in [97])
and may have been known before this. An image of granules
in Azotobacter chroococcum (taken using a transmission
electron microscope) is given in Fig. 2 as an example [98].
The genetically manipulable model organism in PHA
research, Ralstonia eutropha, can accumulate P(3HB) in the
form of multiple granules of 0.2–0.5 ␮m in size, and can
have a polyester content of more than 90% of the cell
dry matter [99], with an average final number of gran-
ules in a typical PHA-producing cell of around 10 [2,100].
Steinbüchel et al. [101] presented some calculations on
P(3HB) in granules: assuming a relative molecular mass of
3 × 106 Da, then a single P(3HB) molecule would be made
up of approximately 35,000 3HB units, with a total length
of 24,000 nm. Based on the typical density of 1.2 g cm−3 ,
the diameter of a granule of a single P(3HB) molecule
would be around 9.7 nm. In a typical granule of 350 nm
diameter, therefore, there would be around 43,000 P(3HB)
molecules. A granule has an amorphous polyester core (see
Section 3.3). A distinct boundary layer at least 3–4 nm thick
was noted in early thin-section electron microscopy and
was thought to comprise a phospholipid monolayer with
402 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
Fig. 2. Transmission electron micrograph of ultrathin section of Azotobac-
ter chroococcum cell treated with phenylacetic acid.
Reprinted from Nuti et al. [98], with permission © Canadian Science Pub-
lishing or its licensors.
embedded and attached proteins; later results suggested
that this layer may be as thick as 14 nm [94,102,103]. Other
results obtained using atomic force microscopy (AFM) on
lysed cells seemed to suggest a smooth outer envelope
(lipid) with an inner network structure ∼2–4 nm thick
based on PhaP (protein) and a crystalline layer beneath this
[104]. However, Western blot and immunogold labeling
experiments have shown that the surface of the granule
has a high protein content. Electron cryotomography has
recently been used [105] to examine granule synthesis and
development in a “near-native”, “frozen-hydrated” state in
intact cells. All had a discontinuous surface layer more con-
sistent with a partial protein coating than a lipid mono- or
bi-layer [105]. It was proposed that the results of earlier
studies may have been in part affected by artifacts of the
preparation methods used as well as differences in nucleoid
condensation.
Granule associated proteins can be divided into 4
groups [3]: the PHA synthases (see Section 3.3), PHA
depolymerases, regulatory proteins (such as PhaR, the tran-
scription factor that negatively regulates PhaP expression
[105,106]) and phasins (PhaP). Phasins are the most abun-
dant protein at the granule surface, and are predicted to
have high (∼90%) proportions of an ␣-helical structure
[107]. Their expression is induced by PHA accumulation
[106], with expression levels increasing and decreasing in
parallel with PHA production [108,109]. They can consti-
tute as much as 5% of the total cellular protein, and have
been variously categorized as constituting between 14 and
54% of the granule surface [107,109], although Neumann
et al. [107] suggest that the coverage may be higher than
this. Each granule has 1–2 molecules of phasin per PHB
chain [109]. Phasins are amphiphilic, with a hydropho-
bic domain exposed to the PHA and a hydrophilic domain
exposed to the cell cytoplasm [3]. They are low molecular
weight proteins (between 11 and 25 kDa), are non-related
and share no sequence homology [110]. Phasins are only
produced during PHA accumulation and are thought to
positively influence PHA accumulation although the mech-
anism is not clear and they are apparently not essential for
PHA accumulation. They are however known to control the
amount [106,108], size and number of granules [111], and
possibly also help prevent PHA crystallization [112]. In the
absence of phasins, cells accumulate only one very large
granule, taking up all available space in the cell [111]. It has
also been shown that overexpression of the phasin leads to
smaller granules [99]. Phasins can bind to any hydropho-
bic inclusion regardless of its composition [3] and appear
to bind to the synthase when conditions are not right for
PHA accumulation. Cells have been found to generally stop
increasing their P(3HB) content at around 80–90% of the
cell dry mass. It is known that PHA synthase activity and
intracellular monomer concentration remain high at this
point, so it is likely that physical constraints are at least
one limiting factor for polymer accumulation [113].
Three models at least have been developed to describe
PHA granule formation in vivo [94,114,115]. One is that
micelles form as the polymer grows and the constituents
of the boundary layer become incorporated as the size
increases [113]. A modified version of this model has
recently been proposed [116] in which the initial nucle-
ating species consists of a PhaC dimer covalently attached
to a PHB chain associated with several phasins that are
attached to maintain solubility. It is hypothesized that sev-
eral of these nucleating species fuse together due to the
hydrophobic patches on the PHB chains, and a soluble gran-
ule precursor is thus formed. In a second model, a synthase
localizes to the inner face of the cytoplasmic membrane
with biosynthesis in the inner membrane space with the
granules eventually budding off [94,115]. In other studies,
nascent PHA granules have been found to localize to the
cell poles or occasionally to the cell poles and to the center
of the cells (i.e. the future cell poles) [117,118]. However,
fluorescence [118] and immunocytochemical localization
studies [119] have shown that PhaC is located randomly
throughout the cytoplasm at the surface of the granule, and
biosynthetic pathways associated with phospholipid pro-
duction are downregulated during PHB production [116].
A third model [109,120] proposes that mediation elements
serve as scaffolds for the initiation of granule formation,
which is incompatible with much of the evidence for the
second model but for which other recent evidence has been
obtained [121]. However, in the recent study by Beeby et al.
[105], no granules were found within the cell membranes
nor were scaffolds observed. Instead, all of the granules
were found towards the center of the cytoplasm. Further
work is required to determine which model or combination
of models is correct.
3.3. Mechanism of polymerization
Various mechanisms of PHA polymerization have been
proposed. It has been shown that Class 1 PHA synthase from
R. eutropha exists in both monomeric and dimeric forms
[122–126]. Artificial priming of the synthase with a trimer
of 3-hydroxybutyryl-CoA shifts the equilibrium towards
the dimeric form and reduces the lag phase for polymer
synthesis, suggesting that the two PHA synthase molecules
combine to form a homodimer with two thiol groups as the
catalytically active sites [100,122–126]. It is also known,
from kinetic studies, that the lag phase is dependent on
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
synthase concentration [116] and that the elongation rate
is much faster than the initiation rate [127,128]. In a recent
publication [116], Cho et al. reported the first isolation of
a Strep-2 tagged PHB synthase (PhaC) directly from native
R. eutropha as opposed to expression from recombinant E.
coli, and found that it unexpectedly copurifies with PHB and
the phasin protein PhaP1 in a soluble form. This enzyme
showed no lag phase in CoA production, unlike expressed
PhaC, and it would seem that it is already primed with poly-
meric PHB (HBn where n is large). It was again found that
the polymerization rate was faster than the initiation rate.
Several independent studies suggest that the synthase and
depolymerase are simultaneously present during polymer
synthesis [109].
It has also been shown [129] that synthases possess a
conserved catalytic triad comprising three residues (cys-
teine, aspartic acid and histidine) and that the cysteine
bonds are involved in covalent catalysis. It is suggested
that the histidine activates the cysteine for nucleophilic
attack on 3-hydroxybutyryl-CoA, while the aspartic acid
may function as a general base catalyst to activate the
hydroxyl group of HBCoA for ester formation [7].
Two mechanisms have been proposed for chain elon-
gation to fit with these observations [130] (Fig. 3). Both
involve chain elongation from acylated cysteine. The first
requires only one active cysteine site, and involves both
covalently and noncovalently bound HBn (CoA) interme-
diates. The second requires that the active site be at the
interface of two PhaC monomers (with PHA synthase being
present in a dimeric form), with the chain being always
covalently attached to one of the two synthase units,
switching between them with the addition of each new
HA unit. In both mechanisms, the cysteine is activated for
nucleophilic attack by a base residue such as histidine in
the active site. The authors found through labeling studies
using the slower synthase mutant C149S-PhaEC that non-
covalently bound species were present in the early stages
of synthesis, and that these intermediates functioned in
a chemically and kinetically competent fashion, lending
weight to the possibility that the first mechanism is more
likely. In another proposal, Yamanaka et al. [131] suggested
that an active site of PHA synthase could be first capped
with succinyl-CoA (formed by the condensation reaction of
3HB-CoA with succinyl thiolate), although the MALDI evi-
dence on which this hypothesis is partly based could have
different interpretations such as a 3HV-terminal unit.
It is well established that microbial PHB (and PHA in
general) is amazingly monodisperse [109]. Doi et al. [132]
estimated the concentration of PHA synthase in a cell (for
R. eutropha) to be 5 ± 1 × 10−7 mol L−1 , making the number
of PHA synthase molecules per cell approximately 18,000
(i.e. a ratio of around 42:1 P(3HB) molecules to P(3HB)
synthase molecules based on Steinbüchel’s result ([101],
Section 3.2)). Tian et al. [109] also found that the ratio of
PHB chains to PhaC class I synthase was ∼60, with 1–2
molecules of PhaP1 (phasin) per chain [109,116]. Hence it
has been supposed that there are many chain termination
and reinitiation events during granule formation [111].
The mechanism of chain transfer or chain termination is
not well understood at this point [105,116]. Kawaguchi and
Doi [133] in the original model for the mechanism of chain
403
termination proposed that a nucleophilic “chain transfer
agent” would attack at the active site thioester, thereby
releasing PHB and terminating elongation [134–138]. Mad-
den et al. [135] found through end group analysis by
31 P NMR of polymer derivatised with 2-chloro-4,4,5,5-
tetramethyl-1,3,2-dioxaphospholane that all groups were
covalently linked to PHA at the carboxyl terminus, and that
all chain transfer agents possessed one or more hydroxyl
groups. These agents may be diol compounds such as ethyl-
ene glycol, polyethylene glycol (PEG) and propylene glycol
as well as 3-hydroxybutyryl-CoA and water; they may
also be enzymes with a water molecule in its active form
[134] or 3-hydroxybutyric acid, which is not enzyme bound
[135]. The carboxyl group at the terminal of the polymer
chain is believed to be esterified with the hydroxyl moi-
ety of the chain transfer agents [137–144]. In support of
this hypothesis, molecular weight is substantially lowered
when various alcohols or poly(ethylene glycol)s are added
to the culture medium [139–141,145]. Tian et al. [109]
also showed that a surface-exposed amino acid was able
to hydrolyze the polyester chain in a chain transfer reac-
tion. Likewise, an increasing concentration of methanol has
been shown to have a large negative effect on the molec-
ular weight of P(3HB) produced by Protomonas extorquens
[145].
However, Lawrence et al. [92] proposed alternative
models, based on the use of an alternative substrate ((R)-3-
hydroxybutyryl-N-acetylcysteamine, HB-NAC). One option
is that chain transfer actually occurs through cleavage of an
internal ester bond within the covalently bound polyester
chain rather than at the thioester linkage; the liberated NAC
functional group could act as chain transfer/termination
agent at either site. Another option is that the noncova-
lently bound polymer intermediate (as in Fig. 3) covalently
linked to NAC could dissociate from the active site. In
addition, it was proposed that the alcohols used for molec-
ular weight control may in fact have been associated with
the granule and then been involved in a transesterifica-
tion reaction during the long fermentation, inducing chain
hydrolysis [92].
It was further noted [92] that there was a probable
change in the system during polymerization (for which
there is yet no explanation) that allows termination only
after several hundred monomers have been incorporated.
Results suggest [130] that as the HB chain grows longer,
the rate of hydrolysis of the covalent linkage to PhaC is
dramatically reduced. The cyclic process of simultaneous
accumulation and turnover of P(3HB) in bacterial cells has
been demonstrated by Doi et al. [146] and Taidi et al. [147]
under nitrogen limitation conditions [148].
In 1986 it was believed that the P(3HB) in granules was
present as a crystalline solid, so the observation (using
solution state NMR) that P(3HB) in live cells was in an
amorphous state was unexpected [149–152]. It was found
that treating the cells to deactivate the granules also led
to a loss of high resolution in the NMR spectrum, with the
simultaneous appearance of crystalline P(3HB), as judged
by X-ray diffraction (XRD). Purified PHAs and even isolated
PHA granules are known to crystallize rapidly under some
conditions. This suggests an important role for the granule
surface layer in preventing crystal nucleation [19,112].
404 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442

Fig. 3. Proposed polymerisation mechanism for the synthesis of PHA.

Reprinted (adapted) with permission from Li et al. [130] Copyright (2009) American Chemical Society.

There are two general hypotheses to explain this result.
One is that the amorphous state of P(3HB) in vivo can be
caused by a simple physical-kinetic limitation [153,154]
within granules based on their small size (stabilized by an
amphiphilic shell). This assumes that the rate of P(3HB)
crystallization in vivo is governed by the rate of sponta-
neous nucleation events, which depends upon the granule
volume and therefore on the third power of the granule
radius. The upper limit for the rate constant of spontaneous
crystal nucleation in isolated P(3HB) at 30 ◦ C (the tempera-
ture of bacterial growth) is 2.5 events mm−3 s−1 [155,156].
The corresponding rate of nucleation for a typical storage
granule of diameter 0.25 ␮m is 2.0 × 10−11 events s−1 . Pro-
vided the granules do not coalesce and are not exposed
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
to foreign nucleating agents, native granules should have
a crystallization half-life of at least 3.4 × 1010 s, or >l000
years [112]. Even at a maximum nucleation rate of less than
100 events mm−3 s−1 , the average lifetime for a granule
before crystallization should still exceed several months.
The in vivo state is therefore proposed to be under
kinetic, not thermodynamic control [112,154]. Horowitz
and Sanders simulated the formation of granules using
synthetic surfactant to coat the surface of submicron-size
artificial granules, which were shown to be amorphous and
stable in suspension, and which had the same size, den-
sity, morphology and molecular mobility as native granules
[112,157].
The other theory is that the water present in the granule
at around 5–10% of the total content could act as a plasti-
cizer, forming hydrogen bonds with the carbonyl groups
of P(3HB) [19]. The proposal is that this hydrogen bonding
prevents strong dipole–dipole interactions from occurring
between the C O and CH3 groups in P(3HB) [19,158,159].
Once this P(3HB)-bound water is removed, in situ crystal-
lization can then occur due to increased hydrogen bonding,
in a similar manner to the crystallization process on cooling
for pure P(3HB) [19,158].
Once released from the granule structure, PHA crys-
tallinity then develops and is core to the overall end-use
properties of PHA including their physical, chemical,
mechanical and rheological characteristics [148].
4. Crystal structures of PHA
In plates and films and similar bulky forms, P(3HB) is
usually a multilamellar system that has been aggregated
into multi-oriented lamellar crystals [160]. Spherulites are
usually formed when P(3HB) chains are crystallized from
the melt in bulk materials; the lamellar crystals grow radi-
ally with twisting. The crystallographic a axis is radial with
b and c axes rotating about it, and as a result the spherulites
usually have a banded texture [155,161]. The patterns of
this banding will change depending both on the crystal-
lization temperature and the molecular weight. Lamellar
thicknesses in spherulites typically range from 4 nm to
10 nm depending on the crystallization temperature [162].
In order to understand the nature of these macrostructures
though, it is important to understand the basic crystal units
from which the lamellae are constructed, and the influence
of comonomer composition on those crystals.
4.1. The ˛-form crystals of PHA
The most common crystal structure of the P(3HB)
molecule is the ␣-form, which is produced under the
typical conditions of melt, cold or solution crystalliza-
tion. In oriented fibers it has been characterized using
X-ray diffraction analysis, and has been shown to adopt
a left-handed 21 helix (G2 T2 )2 (␣-form). The unit cells
are orthorhombic with a space group of P21 21 21 (D42 ) and
with dimensions a = 0.576 nm, b = 1.320 nm and c (the fiber
period) = 0.596 nm [163]. Two molecules of the polyester
pass through the unit cell and the chains are anti-parallel.
Conformational analysis based on potential energy calcu-
lations indicate that the ester groups of the antiparallel
405
chains within the unit cell are nearly at the same level,
and the angles between dipoles was calculated to be
∼60◦ . From this it was concluded that the dipole–dipole
interaction is one of the main factors governing the molec-
ular packing [164]. P(3HV) is also orthorhombic with a
space group of P21 21 21 (D42 ), this time with dimensions
a = 0.592 nm, b = 1.008 nm and c (fiber axis) = 0.556 nm
[165,166]. It also has a twofold screw left-handed sym-
metry along the molecular axis. Crystallographic data for
a range of mcl-PHA copolymers is reported in Sudesh and
Abe [9].
The crystalline morphology of folded single chain crys-
tals has been shown to conform with the classical folding
behavior of synthetic linear aliphatic/thermoplastics; the
crystallographic a-axis runs along the direction of each
crystal and on the crystal surface carboxyl groups and
hydroxyl groups seem to exist alternately [167]. PHA
adopts a helical conformation due to the formation of a
favorable carbonyl oxygen/methyl interaction, unlike pol-
ypropylene for which the driving force is unfavorable
methyl/methyl (van der Waals) interactions [168,169].
Because of this bipolar interaction between continuous
anti-parallel helical chains, stabilized by the ester moieties,
no intramolecular hydrogen bonds are needed to stabi-
lize the helix, unlike proteins [16]. The crystal structure
at the surface of P(3HB) thin films has been studied using
infrared reflectance-absorption and transmission spectra;
the results show that the C H· · ·O C hydrogen bonding of
P(3HB) exists along the a-axis (not the b-axis) between the
CH3 group of one helix and the C O group of another helix
[170].
Single crystals with well-defined structures are
the “monolamellar” form of PHA. Single crystals of
P(3HB) have been prepared from a dilute solution of
polymer in many kinds of organic solvents, such as
chloroform/ethanol [169,171], propylene carbonate
[160,172–174], poly(ethylene glycol) [175], etc. The thick-
ness is dependent on many factors including the molecular
weight, solvent, and crystallization temperature. Typi-
cally P(3HB) forms lathe-shaped single crystals, around
∼5–10 ␮m long and ∼0.3–2 ␮m wide. These single crystals
are usually 4–10 nm thick, depending on the molecular
weight, solvent and crystallization temperature. Solution
grown lamellar crystals of P(3HV) (both synthetic and
bacterial) are square-shaped with a width of 2–4 ␮m and
a thickness of 5–6 nm [168]. From the electron diffraction
diagram, the polymer chains are aligned perpendicular to
the base of the crystal. “The observation of a screw dislo-
cation morphology suggests that the single crystals appear
to be formed by the superposition of several monolamellar
crystals” [168].
Single crystals of P(3HB-co-3HV) have also been pre-
pared [16,167,172,176]. Mitomo et al. [172] looked at single
crystals with six different 3HV contents (at <30% HV). Up
to 10 mol% of HV, the crystals were very similar in mor-
phology to those of P(3HB). However, between 17 and
30 mol% HV, the crystals had morphological irregularities
and a bumpy surface. In P(3HB-co-3HV) spherulites, the
long spacing (Lp ) and lamellar core thickness (lc ) values
are 7–24 nm and 4–17 nm respectively when the (R)-3HV
content is <21 mol%.
406 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442

According to Marchessault et al. [177], mcl-PHAs could crystals in the FT-IR (Fourier-transform infrared) spec-
be classified as thermotropic liquid crystals. The conforma- trum of stretched P(3HB), with bands at 966, 935, 908,
tion was found to be “herringbone” or “comblike” – and it 858, and 1735 cm−1 . For two-step drawn films these peaks
was not found to be possible to distinguish between the two are relatively weak and the crystallinity is small indicat-
possibilities. Polyhydroxyoctanoate (PHO) was found to be ing a disordered structure. These peaks disappeared with
about 25% crystalline. The structure was pseudo-network annealing at higher temperatures.
in character with an initial modulus around 10–20 times The structure of P(3HB-co-3HV) at a clay surface in a
greater than that of a slightly crosslinked rubber. composite is similar to the ␤-form but more ordered, with
chains in a planar zigzag conformation in parallel with
4.2. The ˇ-form crystals of PHA the silica layers. The WAXS reflection (d = 0.480 nm) cor-
responds to the mean distance of P(3HB-co-3HV) chains
The ␤-form of the P(3HB) crystal is a strain-induced in a pseudohexagonal structure. This crystal form disap-
paracrystalline structure of highly extended chains with pears with annealing [188]. Antipov et al. [189] showed
a fiber period of 0.46 nm corresponding to a twisted pla- that when gel spinning was used to obtain highly oriented
nar zigzag conformation [178]. It has an orthorhombic fibers, a third phase (in addition to the crystalline and
crystal system with unit cell parameters of a = 0.576 nm, amorphous) was present and was identified as a columnar
b = 1.320 nm, and c (fiber axis) = 0.598 nm and space group mesophase with 2D-pseudohexagonal packing of the cen-
P21 21 21 . It can be obtained under such conditions as further tral axes of the polymers; it is reversibly strain induced and
drawing of uniaxially stretched films, and is metastable – it is a thermally stable phase formed only under mechanical
can be annealed back to the ␣-form at 130 ◦ C, with a result- stress.
ing increase in crystallinity. The ␤-form was first observed
in a stretched P(3HB) film by Yokouchi et al. [163], and also
4.3. Isodimorphism in P(3HB-co-3HV) copolymers
obtained in uniaxially cold-stretched films of P(3HB) and
P(3HB-co-3HV) [179]. From the calculated value of the fiber
Semicrystalline random copolymers can on rare occa-
repeat length for fully extended P(3HB), the ␤-form has a
sions be isodimorphic in nature, whereby two or more
near completely extended chain conformation. The shift to
comonomer units can exist in the crystal unit structure
the ␤-form did not require prior alignment of the ␣-form
of the other monomer units in the copolymer. As a result
lamellae, but followed orientation of the free chains in the
there is approximately the same high degree of crystallinity
amorphous regions between the crystals [180]. In partic-
across the compositional range, with little resulting dis-
ular, the tie molecules between the lamellar crystals were
ruption to the crystal unit dimensions. There is usually
strongly extended, which leads to high mechanical proper-
pseudoeutectic melting behavior and a structural transi-
ties. Techniques such as melt spinning and drawing of PHA
tion occurs near the pseudoeutectic composition [190,191].
fibers or one- or two-stage cold drawing of films followed
P(3HB-co-3HV) is one such isodimorphic copolymer. A
by annealing have been shown to dramatically increase the
relatively high crystallinity (>50%) has generally been
tensile strength, Young’s modulus and elongation to break
observed across the entire range of HV content (0–95 mol%)
of PHA materials and are associated with the introduc-
in P(3HB-co-3HV) (by XRD); when the materials have been
tion of a ␤-crystalline form as at least part of the lamellar
carefully fractionated and allowed time to fully develop
structure [180]. These properties remain unchanged after
their crystallinity [192] there is evidence of a minimum at
several months suggesting that secondary crystallization
the pseudoeutectic (Fig. 4).
is suppressed when materials are highly orientated and
crystalline [181].
A different mechanism for the formation of the ␤-form
crystals was proposed for one-step-drawn P(3HB-co-3HV)
fibers after isothermal crystallization [182]; it was sug-
gested that many small crystals grow in the amorphous
region during slow isothermal crystallization near the Tg ;
the ␤-structure is then developed by the stretching of
molecular chains in the constrained amorphous region.
Iwata et al. [183] have also reported a 3D crystal structure of
the ␤-structure that has lattice parameters of a = 0.528 nm,
b = 0.920 nm, and c (fiber axis) = 0.469 nm as an ortho-
rhombic crystal system. Etching with enzymes suggests
that the structure of stretched films is stretched chain core
in a shish-kebab morphology with the plates normal to the
stretching direction [184,185].
Nishiyama et al. [186] used a 2D NMR 13 C off-magic-
angle spinning technique to look at the ␣- and ␤-forms of
the P(3HB) crystal. The orientation of the carbonyl carbons
Fig. 4. Effect of 3HV content in P(3HB-co-3HV) on polymer crystallinity
was shown to be in good agreement with the molecu-
by XRD.
lar models proposed by Marchessault and others. Sato Reprinted (adapted) with permission from Wang et al. [201]. Copyright
and coworkers [187] also found evidence for the ␤-form (2001) American Chemical Society.
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
Fig. 5. Effect of comonomer content in PHA on melting temperature, heat
of fusion and glass transition temperature. Plot shows (a) melting temper-
ature (Tm ), (b) heat of fusion (Hm ) and, (c) glass transition temperature
(Tg ) versus content (mol%) of second monomer unit (3HV, 3HP, 3HHx and
4HB) for well fractionated copolyesters; 䊉 P(3HB); : P(3HB-co-3HV); :
P(3HB-co-3HP); : P(3HB-co-3HHx); : P(3HB-co-4HB).
Reprinted (adapted) with permission from Ishida et al. [196], Copyright
(2005) International Union of Pure and Applied Chemistry.
A minimum in the melting temperatures and in the
enthalpy of melting at around 30–55% HV has also been
consistently demonstrated [165,190,193–200] with more
recent literature proposing that it is actually a narrow win-
dow around 47–52% HV once the copolymers are carefully
solvent fractionated [201,202] (Fig. 5).
Melting point depression in polymer mixtures is usu-
ally explained by the Flory equation, but for P(3HB-co-3HV)
this is not applicable since it was derived based on the
assumption that copolymer crystals are composed of only
one type of comonomer component and the other is in
407
the non-crystalline region. This pseudoeutectic has instead
been attributed to the disturbance of crystal packing by
the inclusion of HV units in the P(3HB) lattice and vice
versa [190]. At the pseudoeutectic, the polymer structure
in the crystalline region shifts from a P(3HB) type crys-
tal (fiber repeat 0.596 nm), to a P(3HV) crystal one (fiber
repeat of 0.556 nm) (by XRD), with the middle region orig-
inally being supposed to have both crystal types present
[190]; however later work [201] showed that this was due
to blends being present. It was originally proposed that the
3HV units could sit in the P(3HB) crystal without distorting
the crystalline structure or lattice significantly, and like-
wise for the HB units in the HV crystal. However, solution
grown single crystals of P(3HB) and P(3HB-co-3HV) (up to
30% HV) showed a linear increase in melting points and a
hyperbolic increase in long spacings as the crystallization
temperature increases [172]. As the HV content increases,
the lattice indexes for the HB crystal expand up to the max-
imum (a = 0.602 nm, b = 1.343 nm and c = 0.604 nm) and the
crystal is particularly elongated in the a and b parame-
ters; the d spacing in the HB crystal lattice increases. By
contrast, the HV crystal structure slightly contracts in the
b axis but essentially remains unchanged throughout the
range [165,190,194,201,203], so that the HB unit can sit
in the HV crystal lattice without distorting it. Iwata et al.
[204] prepared single crystals of P(3HB-co-8 mol% 3HV)
and analyzed their electron diffraction diagrams; it was
found that the d-spacings for the P(3HB) homopolymer
and the P(3HB-co-8 mol% 3HV) copolymer were the same,
implying that (R)-3HV units are excluded from the crys-
tals and exist mainly on the crystal surfaces during slow
crystallization, with a small proportion acting as a defect
group in the crystal. In this case, the ethyl side chains of
the HV component cause the (1 1 0) plane of the HB crystal
to expand due to steric effects [172,192]. These results are
confirmed by other studies. By solid state NMR, for exam-
ple, it has been shown that at low HV contents the crystal
almost exclusively comprised the HB crystal unit with HV
effectively excluded, and the less ordered regions being
enriched in HV, and at higher levels of 3HV content the
3HV crystal structure dominated. A plot of mol% 3HV in the
crystal and amorphous phase was produced (Fig. 6). At 47%
HV, the less ordered portion of HV increases on stretching
while that for HB disappears. On annealing this behavior
is reversed [153,190]. This work was confirmed by other
NMR studies using very carefully fractionated copolymers,
which showed that the HB units are easier to crystallize in
the HV unit than vice versa [205]. The 3HV units are mostly
excluded from the crystalline region if the HV content is
<20% [205,206].
Density measurements also support this proposal. The
density of the amorphous PHA is 1.177 g cm−3 and of the
crystalline phase is 1.260 g cm−3 [160]. Barker et al. [207]
showed that the density of P(3HB-co-3HV) in both the
crystalline and amorphous states decreased as HV content
increased. Assuming that the degree of crystallinity of the
copolymers of different HV content is similar, then there
must be partial exclusion of HV units from the crystals
[207]. Mitomo and Doi [195] isolated the crystalline core
using aminolysis (with 20% methylamine), and showed
that the HV content of the core was smaller than the
408 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
Fig. 6. Composition dependence of HV content of P(3HB-co-3HV) in the
crystalline phase; 䊉 HV content of crystalline phase;  HV content of
amorphous phase.
Reprinted with permission from Bonthrone et al. [154], Copyright (1992)
Wiley-Blackwell.
overall content in P(3HB-co-3HV) copolymer, indicating
some exclusion of HV.
By using compositionally fractionated and labeled
copolymers, Yoshie et al. [208] also showed there is a struc-
tural transition in the crystal state at around 10 mol% HV. It
was confirmed that above 10 mol% HV the d(1 1 0) spacing
increased with an increase in HV, and a small increase in
d(0 2 0) was also observed. The d(0 0 2) spacing remained
constant, confirming the expansion in the ab plane. Below
10 mol% HV, the crystal was the same as for P(3HB). Two
models for P(3HB-co-3HV) crystals were introduced: (1)
the sandwich model, which applies below 10 mol% HV: in
this case the core of the lamella is pure HB units, and the
HV units are at the crystal edges and in the amorphous
regions; and (2) the uniform lamella model in which the
HV units are uniformly distributed through the crystal. It
was of note that crystallization rates dropped rapidly above
10 mol% HV as well.
From Molecular Mechanics (MM2) modeling [191], the
incorporation of the 3HV unit into the P(3HB) type lattice
has a substantial influence. This is attributed to the steric
interaction between the ethyl side group and the nearest
carboxyl oxygen atom. However, it was also found that
replacement of a 3HB unit by a 3HV unit only affected the
conformational angles of one of the nearest neighboring HB
units; therefore the overall 21 helical structure remained
unchanged. The effect of crystallization temperature on
the extent of co-crystallization has also been investigated
[156,209,210]. The 3HV content in the P(3HB) lattice has
been shown to decrease as the crystallization temperature
increases; likewise at higher HV contents when the P(3HV)
lattice is present, the 3HB content decreases with tem-
perature. Barham and coworkers [182] pointed out that
the mechanism for inclusion of comonomer units into the
P(3HB) crystal unit is by kinetic (i.e. dependent on a combi-
nation of time and temperature), rather than equilibrium,
methods. According to Yoshie et al. [212] co-crystallization
results from the entrapment of P(3HB-co-3HV) chains
near the growing front of the crystals. The extent of
co-crystallization therefore decreases with an increase in
crystallization temperature because more arrangement of
the chain can take place during crystallization at the higher
temperature, providing the opportunity to approach the
equilibrium state. The temperature dependence of the
crystal structure and the weak intermolecular interactions
in P(3HB-co-3HV) has also been analyzed using FT-IR [213].
It was found that for HB-rich copolymers, the CH3 · · ·O C
hydrogen bond between the CH3 group and the C O
group is found only in the HB part of the crystal structure,
while the high HV content materials bond through the
CH2 · · ·O C groups. For P(3HB-co-3HV) up to 21 mol% HV,
there was a sudden drop in crystallinity above 170 ◦ C. By
contrast, for 28.8 mol% HV, the change began much earlier
at around 70 ◦ C. The P(3HV) homopolymer lost crystallinity
earlier and at lower temperatures than P(3HB), and at
the eutectic (HV = 58.4 mol%) crystallinity began to be
lost even at room temperature. WAXD spectra were also
followed with increasing temperature and showed similar
results. It was proposed that the lamellar structure in
these copolymers deformed with increasing temperature,
for steric reasons, and that these weak hydrogen bonds
are very important for the stabilization of the lamellae. In
addition it has been found that there is a rigid component
in the non-crystalline region of P(3HB) and P(3HB-co-3HV)
at room temperature, which decreases with increasing
temperature and HV content [214]. This may be due to the
tightly packed structures adjacent to the lamellae.
When the 3HV comonomers are included in the P(3HB)-
type lattice, they will result in “less-perfect” crystals with
more defects, smaller crystalline domains and less brit-
tleness [215]. At higher HV contents the spherulites are
small as the rate of crystallization is slower and/or there are
relatively more nucleation sites. Less perfection in the crys-
talline domain in turn corresponds to films with greater
flexibility and toughness [215].
By contrast, none of the other comonomer units found
in PHA have demonstrated any degree of co-crystallization.
Mitomo and Doi [195] found that there was no change in
lattice parameters in P(3HB-co-4HB) with increasing 4HB
content, implying that there was no inclusion of 4HB in
the 3HB crystal, and Iawata et al. [204] also demonstrated
that the second monomer units for P(3HB-co-6HHx) and
P(3HB-co-4HB) copolymers are largely found on the crys-
tal surfaces (and also that these surfaces are irregular with
loose loop chain foldings). Xie et al. [216] likewise found
that HHx is excluded from the HB crystal (at 7.2% HHx),
and high-resolution solid-state 13 C NMR of composition-
ally fractionated bacterial P(3HB-co-3HP) [217] showed
that co-crystallization did not occur in the P(3HB-co-3HP)
system. From the lineshape analyzes of the 3HB methyl car-
bon peak it was also shown that the presence of minor 3HP
comonomer units significantly suppressed the crystalliza-
tion of a 3HB-rich semicrystalline copolyester; there was a
correlation between HB block length and the crystallizabil-
ity of the copolymers.
The glass transition temperature does not show pseu-
doeutectic behavior, as it is reflective of molecular motion
in the non-crystalline phase; instead it shows a linear rela-
tionship, decreasing with an increase in comonomer unit
content [197] (Fig. 5). The Tg of P(3HA)s with long side
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
chains is about −40 ◦ C, which decreases with an increase
in the average side chain length [218].
5. Notes on the analysis of crystallinity in PHA
copolymers
With respect to the previous discussion on composition,
crystallinity and crystal structure, there are notes of caution
to be made. The analysis of crystallinity is not straightfor-
ward for PHA copolymers.
Crystallinity can be expressed in three ways depend-
ing on the analytical approach, and can include the volume
fraction of crystal, the mass fraction of crystal, or the molar
fraction of crystal [207]. WAXD methods of analysis mea-
sure the intermolecular order as a result of chain packing
(effectively the mass fraction crystallinity) and are the
method of choice for determining relative crystallinity of
these materials, particularly if the Ruland method (which
requires diffraction data over a wide angular range and
correction for air scattering, incoherent scattering, and
diffuse scattering arising from thermal fluctuations and
paracrystalline disorder) is used [218]. A number of other
techniques have also been used over the years, and most
give results that are consistent with XRD results. For
example, by 13 C NMR analysis using single-pulse excita-
tion with magic-angle spinning, the crystallinity of P(3HB)
and P(3HB-co-3HV) (2.7 and 6.5 mol% 3HV) was found to
be 68%, 60% and 56% respectively; this compared well
with WAXD data of 60, 55, and 53% [219]. Similarly, in
another NMR-based study it was found that the degree
of crystallinity was 60%, 60%, 58% and 56% for P(3HB) and
P(3HB-co-3HV) (at 4.33, 6.96 and 10.5 mol% 3HV respec-
tively) [214]. Kansiz et al. [220] found by FT-IR that
P(3HB-co-3HV) (14 mol% HV) had 47% crystallinity at full
development of crystallinity (which was slightly lower
than found by the other techniques but still in line) – this
method measures the molar fraction of crystals and takes
into account short-range order (such as in tie molecules)
[207].
Differential scanning calorimetry (DSC) is an extremely
useful tool for the characterization of PHA materials, par-
ticularly in light of the significance of rate of crystallization
with respect to polymer properties and also of the impact
of melt temperature on polymer processing, thermal sta-
bility, etc. The significant features of the trace include the
melt temperature (Tm ) and the heat of fusion (Hf ), the
crystallization temperature (Tc ), the cold crystallization
temperature (Tcc ), and the glass transition temperature
(Tg ). However, the use of DSC for the measurement of crys-
tallinity in PHA copolymers based on heat of fusion needs to
be approached with much care. According to Barham et al.
[211], “the measurement and definition of crystallinity is
not a simple matter even in homopolymers; in copolymers
it becomes very complex”. While there is adequate agree-
ment between the results obtained for different methods
such as XRD and DSC for the P(3HB) homopolymer, in PHBV
copolymers, “it is quite difficult to measure a true crys-
tallinity, since the density and heat of fusion measurements
are affected by the partitioning of the HV units between the
crystalline and amorphous regions and can only be used for
409
the extreme cases of complete inclusion or exclusion of the
comonomers from the crystals” [211] – see also [207].
Melting temperatures and the enthalpy of melting are
known to be dependent on the thickness of the crystalline
lamellae. This thickness varies with crystallization condi-
tions. Therefore the melting temperature as detected in
the DSC depends on crystallization conditions such as crys-
tallization temperature, crystallization time and annealing
treatment.
One of the standard methods for determining the crys-
tallinity (X) of semi-crystalline polymers is by comparing
the area of the melting peak (Hf ) with the melting
enthalpy of a 100% crystalline material (Hf 100% ):
Hf
X=
Hf 100%
An estimate of the heat of fusion of an infinite crystal
of P(3HB) was obtained using a slowly annealed film of
86% crystallinity and was found to be 146 J g−1 by den-
sity measurement [160], which is the usual value used in
this calculation. Bauer and Owen [221] obtained a higher
value of 170 J g−1 for P(3HB). Steinbüchel and Schmack
[222] found that P(3HV) had a Tm of 108 ◦ C with H
of around 91 J g−1 . Pearce and Marchessault [223] used
Hf 100% for 95 mol% HV of 128 J g−1 (from [165]) and for
100 mol% P(3HV), 131 J g−1 (from [224]). For other PHBV
copolymers of lower HV content, a common value used for
Hf 100% is 109 J g−1 (obtained from WAXS and DSC analy-
sis of PHBV containing 10 mol% 3HV) [225]. However, some
authors have used slightly modified techniques to calcu-
late percent crystallinity in these copolymers. Cheng and
Sun [226] took Hf 100% as 146 * mol fraction HB J g−1 ; this
was used since at low HV content (<10%) the sequences of
HV units are excluded from the lamella. By contrast Chan
et al. [227], in blends of P(3HB-co-3HV) (12 mol% 3HV), took
H = HPHB * wt% PHB + HPHBV * (1 − wt% PHB), where
HPHB = 146 J g−1 and HPHBV = 109 J g−1 . An alternative
measure for Hf 100% for P(3HB-co-3HV) = 3427 cal mol−1
(or 14.3 kJ mol−1 ) of crystalline repeat unit according to
Orts et al. [228]. (Values given in kJ mol−1 are not con-
verted into J g−1 given the variation in molecular weight
depending on repeat unit composition.) Kamiya et al. [205]
used Hf 100% for P(3HB) = 11.2 kJ mol−1 (Tm = 175 ◦ C) while
Hf 100% for P(3HV) = 10 kJ mol−1 (Tm = 108 ◦ C). Specific val-
ues of H◦ for a range of P(3HB-co-3HV) copolymers were
also obtained using a combination of heat of fusion from
DSC and crystallinity from XRD [165], with values ranging
from a maximum of 130 J g−1 for P(3HB) to a minimum of
92 J g−1 for 34 mol% 3HV. Most of these methods bring the
crystallinities as assessed by DSC more in line with XRD
results; however, the enthalpy of fusion is often still lower
than expected (see Fig. 5 for the enthalpy of fusion of frac-
tionated samples, and data in [203]).
The major uncertainty of methods such as DSC is that
some knowledge of the values of a given property (e.g.,
density or heat of fusion) of the 100% crystalline polymer
must be assumed. Furthermore it is usually assumed that
this property varies linearly with both the degree of crys-
tallinity and the copolymer composition. This is invalid
where co-crystallization occurs [228]. “The borderline of
410 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
copolymer composition between isomorphism and lattice
defects is ambiguous.” [229].
Orts et al. [228] compared the crystallinity obtained
using inverse gas chromatography with the results
obtained from XRD and DSC. In this method, a column was
packed with diatomaceous silica coated with P(3HB-co-
3HV) (7, 21 and 27% HV), and the retention of a “probe”
chemical (decane) was tested as a function of tempera-
ture. Crystallinities of around 60% for all copolymers were
observed, which was consistent with XRD results. By con-
trast, the crystallinities determined from the heat of fusion
showed a large drop from 59% to 24%. It was found that the
decrease in the heat of fusion with increasing HV content
was not describing a decrease in crystallinity but instead
was reflective of a decrease in crystallite size and/or crys-
tal perfection. It was also noted that as the HV content
increases, melting begins at a lower temperature, getting
increasingly close to room temp. This is likely due to sam-
ples of lower HV content crystallizing more readily into
highly ordered crystallites with a relatively large crystal
size. It is also likely that the inclusion of this “other” co-
unit in the P(3HB) crystalline phase systematically affects
the heat of fusion due to defect formation.
Chen and Hwang [230] have shown that crystallinity
obtained by Hf /Hf 100% can be called the ‘equivalent
weight crystallinity’, signifying “the amount of perfect
crystals that can be melted by the measured enthalpy of
melting. The equivalent weight crystallinity does not only
depend on the amount of the crystals, but also on the per-
fection of the crystals: a crystal with higher perfection has
larger equivalent weight” [230]. Samples of P(3HB-co-3HV)
crystals (of 17 and 30 mol% 3HV) grown isothermally from
solution have been shown to be very small and irregular
[172]. The fold surface also became more irregular with
an increase in HV content, which was attributed to inho-
mogeneous distribution of HV units in the crystalline and
amorphous phases. It is also known that lamellar thick-
ening with increased melting temperatures and spherulite
size has been observed when PHA copolymers are crystal-
lized at higher temperatures [162,195,231].
Poley et al. [232] in an alternative explanation proposed
that the difference may be due to a decrease in the amor-
phous phase density. In addition, the presence of air in
the free volume of the polymers may be a consideration;
given the results of oxygen and carbon dioxide mass dif-
fusion, they assumed that the air solubility of the samples
grows with increasing HV. As air-specific heat capacity is
1.17 × 10−3 J cm−3 K−1 , three orders of magnitude lower
than the values in the typical range of polymers, there-
fore it was suggested that air could under these conditions
contribute to a stronger reduction in heat capacity values.
It is important to distinguish multiple peaks that are
due to phase-separated structures from those that are a
result of annealing during heating in the DSC. This can be
done by changing the rate of heating [233]. The possibil-
ities contributing to double or multiple melting behavior
include: (1) melting, recrystallization, and remelting
during heating, (2) the presence of more than one crystal
modifications (polymorphism), (3) different morphologies
(lamellar thickness, distribution, perfection or stability), (4)
physical aging or/and relaxation of the rigid amorphous
fraction, (5) different molecular weight species [234]
and (6) orientation effects, and so on [234–236].
Melt/recrystallization melting peaks are typically due
to the primary crystals formed in crystallization process
(the lower peak) with higher melting peaks formed due to
crystals that have recrystallized during heating.
As a further note, it has been demonstrated that the rate
of crystallization of pure P(3HB-co-3HV) copolymers of 18
and 30 mol% HV is extraordinarily slow and can take weeks
to fully develop [172]. In any DSC studies of blends, not pure
random copolymers, it is important to take into account the
fact that preferential crystallization of the faster crystalliz-
ing phase will exclude the separate, slower crystallizing
components into a separate phase where, with time, these
components can also slowly crystallize. Without allowing
sufficient time (or the best temperature for annealing of the
materials) the DSC analysis will not be reflective of the final
polymer crystallinity. This is also true for pure copolymers
of high HV content.
The increased segmental mobility in the amorphous
phase for P(3HB-co-3HV) is responsible for the observed
decrease in Tg [196]. However, there is a very low enthalpy
for the glass transition of P(3HB) of 0.43 J ◦ C−1 g−1 as
reported in Scandola et al. [237], so as the percentage
of the P(3HB) crystal phase increases, so the Tg for this
phase becomes increasingly difficult to observe, and DMTA
becomes a more reliable method of analysis for this param-
eter.
6. Crystallization kinetics
The crystallization kinetics of PHA polymers are very
important to understand, as one of the potentially limiting
factors for the commercial use of these materials is the
slow rate of crystallization for some of the copolymers.
There are three types of crystallization behaviors (regimes
I–III) according to the Hoffman theory of crystallization
[238]. In regime I (at low supercoolings i.e. at a temper-
ature that is not much below the melting temperature),
the radial growth rate is large relative to the nucleation
rate: growth following each nucleation event is rapid and
complete before the next nucleation event. In regime II,
at lower temperatures, the nucleation rate and the radial
growth rate are comparable, so that both the formation of
new nuclei and the growth of crystals take place together.
In regime III (at high supercoolings), the nucleation rate is
very large in comparison with the growth rate so that there
is little distance between the nuclei.
6.1. Equilibrium melting temperature
The equilibrium melting temperature, Tm 0 , is defined as
“the melting temperature of an infinite stack of extended
chain crystals, large in directions perpendicular to the chain
axis and where the chain ends have established an equi-
librium state of pairing” [239]. This parameter is “one of
the most important thermodynamic properties of crystal-
lizable polymers, as it is the reference temperature from
which the driving force for crystallization is defined” [239].
According to the Hoffman–Weeks method, the equilib-
rium melting temperature Tm 0 is obtained by extrapolating
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
a linear regression of the observed melting temperature
Tm of crystals formed at a range of crystallization tem-
0 values have
peratures (Tc ) to the line Tm = Tc . Various Tm
been calculated for P(3HB) and its copolymers. Tm 0 was
estimated to be 197 ◦ C for P(3HB) and 143 ◦ C for P(3HB-co-
3HV) (30 mol% 3HV) by Mitomo et al. [172], and was found
to be 195 ± 5 ◦ C as judged by one method, 200 ◦ C by another
and 197 ◦ C by a third according to Barham et al. [160]. Organ
[240] obtained a range of 176 ≤ Tm 0 ≤ 188 ◦ C for P(3HB-
co-3HV) (4 mol% 3HV), compared with 179 ≤ Tm 0 ≤ 198 ◦ C
for P(3HB), depending on molecular weight. Other authors
also report values within this range [227]. The equilib-
rium melting temperature for P(3HB) was found to have
a strong and unexpected dependence on molecular weight
[240,241]. For 8 mol% HV, Tm 0 decreases as molecular weight
decreases (from 188 ◦ C to 176 ◦ C as molecular weight falls
from 220,000 to 56,000). The behavior is more complex
for the 18 mol% HV material, suggesting that the degree of
exclusion of HV from the crystal lattice varies with molecu-
lar weight. There was a hypothesis that there is a different,
higher melting crystal structure in polymers above 200,000
molecular weight; however, this was not supported by XRD
[240,241].
The Tm0 of P(3HV) was found to be 130 ◦ C, with a T
g
value of −16 ◦ C (suggesting that the previously reported
−11 ◦ C was due to the partial crystallization of P(3HV) on
supercooling) [223].
6.2. PHA crystal growth kinetics
The kinetics of the growth rate of the P(3HB) spherulite
has been evaluated at various crystallization temperatures
and shows a complex curve against temperature and a
maximum value of about 3–4 ␮m s−1 at ∼90 ◦ C [160]. It
has been reported that the regime of crystallization for
P(3HB-co-3HV) and its blends is assigned to be regime III
(j = 4) when Tc is below 130 ◦ C [156], with the transition
from regime II to regime III also occurring at ∼130 ◦ C for
P(3HB) [160]. Regime I growth has yet to be observed in
P(3HB) [9]. P(3HB) can undergo heterogeneous nucleation
without addition of nucleating agents provided it has com-
pletely melted. The nucleation rate is obtained by counting
the number of spherulites formed during the isothermal
crystallization process. The maximum rate of nucleation
occurs at a lower temperature than that of crystal growth
rate for P(3HB), and in the pure polymer it is sporadic
[155,156,211]. Therefore the overall crystallization rate for
P(3HB) is at a maximum at between 60 and 90 ◦ C [242].
The spherulitic growth rates for P(3HB-co-3HV) copoly-
mers are markedly reduced as the comonomer content
increases to the pseudoeutectic (at ∼50 mol% 3HV)
[165,192,211,217]. Crystallization curves are shifted to
lower temperature, and nucleation rates also decrease,
with the overall rate being some four orders of magnitude
slower at the pseudoeutectic. Above the pseudoeutec-
tic, the rate again increases; at 83 mol% 3HV it is about
the same as for P(3HB) [165]. For example, Bloember-
gen et al. [243] followed melt quenched P(3HB-co-3HV)
(20 mol% HV) films by WAXD (Fig. 7), which was still
amorphous after 15 min and took days to fully crystal-
lize. The ultimate crystallinity was around 65% [243]. A
411
Fig. 7. Variation of crystallisation rate for a range of melt-quenched
P(3HB-co-3HV) copolymers as followed by XRD. 䊉 0% HV;  4% HV; 
7% HV;  8% HV;  17% HV;  20% HV.
Reprinted (adapted) with permission from Bloembergen et al. [243], Copy-
right (1986) American Chemical Society.
similar result was obtained by Kunioka et al. [192]. This
slow rate was attributed to the increasing difficulty for
either crystal lattice to accommodate increasing amounts
of guest monomer units during cocrystallization [165]; the
(R)-HV units have to be regarded as energetically unfavor-
able defects in the P(3HB) crystalline phase and vice versa.
It was found [243] that the crystallization rate of P(3HB-
co-3HV) copolymers from the melt was faster than that
of solution-cast samples, and it was much faster still from
solution-precipitated samples (as analyzed using FT-IR and
XRD). The variation of composition in the crystalline phase
with temperature is also more pronounced in the copoly-
mers with mid-level HV contents. Bloembergen et al. [243]
noted that these materials have an induction period, and
there was a suggestion that the initial formation of highly
ordered crystals was followed by the formation of much
less perfect crystals with lower melting temperatures. It
was found that the crystallite size as judged by the width
at half height of the crystalline peaks (2 ≈ 17◦ ) remained
constant for melt quenched films, but for solution cast sam-
ples the initial thickness was higher but fell with time.
This implies that crystallization is mostly nucleation rate
limited in the first phase, then crystal growth rate limited
in the second.
It has been suggested that CH· · ·O hydrogen bonding
between the C O and CH3 groups in P(3HB) helps to
drive crystal formation [244,245]. During crystallization,
the P(3HB) molecules become constrained in their move-
ment due to close packing into a crystalline state. When the
crystals melt, the CH· · ·O hydrogen bonds weaken, increas-
ing molecular mobility.
In P(3HB-co-4HB), the rate of crystallization also
decreases with increasing comonomer content [192,246].
The crystallization of mcl-P(3HA)s with long side-chains
is very slow [247,248], taking days to weeks to reach full
crystallinity, with an equilibrium crystallinity of about 25%,
which is relatively small compared with that of P(3HB-co-
3HV) [249]. Abe et al. [88] also looked at isotactic/atactic
blends; again the radial growth rate decreased as isotactic
412 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
P(3HB) content increased. The SAXS data showed that the
lamellar periodicity of P(3HB) increased also. The decrease
in crystallinity was fully accounted for by the amorphous
addition of the atactic component, i.e. it did not affect
the crystallinity of the crystalline component, implying
that the atactic P(3HB) is found in the amorphous regions
between the lamellae.
Detailed studies have shown that the high degree of
crystallization and relatively rapid crystallization rate of
P(3HB) generates pores and protrusions on the P(3HB)
film surface, which could prohibit mammalian cell growth
[250]. The presence of P(3HB-co-3HHx) in P(3HB) polymer
strongly reduced both the degree of crystallization and the
crystallization rate of P(3HB). The low degree of crystalliza-
tion of P(3HB-co-3HHx)/P(3HB) blends provided films with
a fairly regular and smooth surface.
6.3. Analysis of crystallization kinetics and mechanism
The Avrami equation describes the overall crystalliza-
tion rate under isothermal conditions, assuming that the
relative degree of crystallization Xt is developed with crys-
tallization time t:
nA
1 − Xt = exp(−kA t ) (1)
which is the double-logarithmic form of
ln[− ln(1 − Xt )] = ln kA + nA ln t (2)
where kA is the crystallization rate constant (min−1 )
and is temperature dependent; nA , an integral, is the
Avrami index, and contains information about nucleation
and growth geometry. Although Avrami analysis can also
be used to describe the non-isothermal processes, the
obtained values of kA and nA do not have the same physi-
cal meaning as in isothermal processes due to the fact that
under nonisothermal conditions the temperature changes
constantly. There have been a range of values obtained for
the Avrami index, nA . For example, Chanprateep et al. [251]
and Chua et al. [252] obtained values for P(3HB) of the order
of 3, and that of P(3HB-co-3HV) was found to be between
2.5 and 3 (for 2.6 mol% and 8 mol% HV) [253,254]. In general,
nA = 3 corresponds to two different possible crystallization
mechanisms. One is three-dimensional growth and instan-
taneous nucleation; the other is two-dimensional growth
and homogeneous nucleation [254]. By contrast, Gunaratne
and Schanks [234] obtained nA ∼ 2 for commercial P(3HB)
and P(3HB-co-3HV) (5–12 mol% HV) which corresponds to
2D crystal growth and heterogeneous nucleation. Liu et al.
[255] also obtained nA values of 2.0–2.2 for P(3HB-co-3HV)
(6.6 mol% HV). Chan et al. [227], however, obtained nA val-
ues across a range of HV contents of 2.35–2.7 and Saad et al.
[256] obtained an exponent of 3.8 (consistent with sporadic
3D spherulitic growth) for thin P(3HB) films. Meanwhile
Xu et al. [257] applied the Avrami equation to IR data for
isothermal crystallization of P(3HB) and Nodax (P(3HB-
co-3HHx)) and obtained an nA value of 1.72 and 2.08
respectively, indicating that a heterogeneous nucleation
mechanism exists in this case. There are other examples for
blends, but there would seem to be considerable variability
in the isothermal crystallization kinetics for P(3HB/V) and
similar materials. Since the crystallization kinetics strongly
influences the end-use properties, it is important to under-
stand the source of this variability, which may be a result
of inherent uncertainties in the method but may also be a
result of variability in the properties of the materials being
analyzed, such as chemical compositional distribution and
levels of impurities.
The activation energy of crystallization was found to be
110 ± 14 kJ mol−1 for P(3HB-co-3HV) (8 mol% HV), larger
than that of P(3HB) [254], meaning that the crystallization
ability of P(3HB-co-3HV) is lower than that of P(3HB). The
work of chain-folding was also lower than that for P(3HB),
implying that the chains of P(3HB-co-3HV) were more flex-
ible.
Non-isothermal crystallization kinetics can also be ana-
lyzed by looking at the relationship between the relative
degree of crystallinity (Xt ) as a function of temperature
at different cooling rates. Ziaee and Supaphol [258] have
reviewed the kinetics of P(3HB) crystallization under non-
isothermal conditions, and found that the Tobin and Ozawa
methods gave comparable results to the Avrami method
under these conditions.
6.4. Effect of blends on crystallization rates
Finally, the crystallization kinetics of PHA copolymers
can be strongly affected by the presence of blended mate-
rials. According to Kammer et al. [259], the sensitivity
of conventional DSC (Tg ) to heterogeneities in polymeric
materials is approximately 50 nm domain size, although
others say that it can provide information on miscibility
and phase structure on the 10–30 nm scale [260,261]. The
crystallization kinetics can be monitored through the crys-
tallization exotherm, with the initial slope being related to
the nucleation rate. If this crystallization rate is depressed,
then this is typically due to (1) a reduction of the driving
force for crystallization due to the changes in the equilib-
rium melting point, (2) a dilution effect (i.e. a change in
the composition of the not-yet-crystallized melt, which is
enriched in components that do not crystallize as rapidly
as those in the developing crystal), which lowers the
probability of a critical nucleus forming in front of the
growing spherulites and (3) entropic factors that produce
an increase in the energy related to the formation of a
nucleus of a critical size [262].
7. Molecular weight control
The molecular weight distribution is also, to a
large extent, responsible for the end-use properties of
biopolymers, via its control of macromolecular and supra-
macromolecular structures. It is known that the molecular
weight of PHA synthesized by biological means is much
higher than that achieved chemically [12]. The mechani-
cal properties of PHA deteriorate when the weight average
molecular weight (Mw ) is lower than 0.4 × 106 Da [263],
and for thermoplastic applications the value of Mw should
be higher than ∼0.6 × 106 Da [264]. Molecular weights typ-
ically range between 0.2 × 106 and 3 × 106 Da [39], and
different bacteria are known to produce P(3HB) of different
molecular weights [71]. Substrate type and concentration,
nutrient availability and growth conditions such as pH
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
and temperature are also important [265–267], and val-
ues of Mn as high as 2 × 107 Da have also been reported
in mutant strains [134]. There is also a low molecular
weight form of P(3HB) (of ∼13,000 Da) which is commonly
found in bacterial cells and is a constituent of the cyto-
plasm and cytoplasm membrane, where it may be part of a
P(3HB)/calcium polyphosphate channel [268].
Accurate determination of the true molecular weight
requires solution viscosity measurement. However, the
Mark–Houwink–Sakurada parameters (which relate the
intrinsic viscosity to the molecular weight) have only
been reported for the P(3HB) homopolymer [269,270]. In
general, the molecular weight data for PHA samples are
obtained for comparison purposes by means of GPC anal-
ysis, where polystyrene standards are used to construct a
calibration curve.
A number of integrated metabolic/polymerization mod-
els have been developed for the prediction of P(3HB)
concentration and molecular weight distribution in bacte-
rial cultures [271–276]. Some models have assumed chain
transfer and polymer degradation reactions to be neg-
ligible, which does not hold true in wild-type bacteria
[148]. More recent approaches have attempted to integrate
current metabolic models containing a realistic descrip-
tion of cell metabolism (i.e. monomer concentration is not
assumed to be constant) with a polymerization kinetic
model comprising initiation, propagation, chain trans-
fer and degradation reactions [124,131,135,274,276,277].
Most models also assume that the concentrations of
polymerase, chain transfer and depolymerase molecules
remain constant during the polymer synthesis, and that the
selectivity of the various P(3HA) polymerases for different
monomers does not change [148].
It has been calculated [132] that approximately two
molecules of (R)-3-hydroxybutyryl-CoA are added every 1 s
into a propagating chain of P(3HB), and that the average
lifetime of a growing chain should therefore be about 1 h
at 30 ◦ C. It has also been shown that the Mn of P(3HB) poly-
mers increases rapidly at first then can decrease with time
[100,278]. Shimizu et al. [279] likewise found that in Alca-
ligenes eutrophus H16 (ATCC 17699) fed with butyric acid,
the maximum average molecular weight of P(3HB) was
already reached within 5 h and was found to be lower when
the butyric acid concentration was increased. Hiraishi et al.
[280] used AFM to study granule formation in vitro on
graphite; in the initial stages, flexible fibrillar P(3HB) chains
(0.3 ␮m diameter) were shown to be growing from the
immobilized synthase particles, and 150–400 nm globules
were formed within 5 min, which then coalesced to form
large granules.
The polydispersity (Mw /Mn ) meanwhile increases (typi-
cally from 1.5 to 2.0) during accumulation [100,278]. These
results could be due to either chain transfer reactions or
depolymerase activity during granule formation or to the
random decay of active synthase molecules [278,281].
The cyclic nature of P(3HB) metabolism can also be used
to manipulate molecular weight – for example Shimizu
et al. [279] found that, in the presence of a nitrogen
source, P(3HB) production and degradation in Alcaligenes
eutrophus took place simultaneously, with the lower
molecular weight P(3HB) being degraded fastest resulting
413
in an overall increase in the average molecular weight. It
is believed to be the balance between these processes that
affects the length of the polymer chain [19].
There have been two approaches to the control of
the molecular weight of PHA through polymer synthesis
to date. The first involves control of the level and time
of expression of PHA synthase [282]. A lower molecular
weight is obtained when high levels of synthase expression
are induced; likewise a higher molecular weight results
when low levels of synthase are expressed [282,283].
However, the molecular weight of P(3HB) produced by
recombinant R. eutropha was found to be independent of
PHA synthase activity [136]. The second method of control
of PHA molar mass involves the addition of poly(ethylene
glycol) (PEG) to the fermentation medium (see Section 3.3
for a discussion of the effect of water or alcohol addition
on the molecular weight of PHA chains). PEG is a neu-
tral, water-soluble polyether that is relatively nontoxic to
cellular systems [88], and associates with membrane phos-
pholipid head groups [284,285].
In other studies, it has been shown that a potassium
deficiency can lead to the biosynthesis of a high Mw P(3HB)
(>3 × 106 Da) in a methane-utilizing mixed culture [286]. It
has also been shown that as the carbon source concentra-
tion increases, so the average Mw decreases [134,145,287].
In summary, since molecular weight is one of the pri-
mary drivers controlling end-use properties, it is evident
that the mechanisms controlling molecular weight need
to be well-defined and tools for modeling and/or control-
ling molecular weight need to be further developed and
integrated into PHA production processes.
8. Control of monomer distribution and
microstructure
Another approach to the modification of chemome-
chanical properties of PHA copolymers is to manipulate the
sequences of monomer units within the polymer chain to
produce non-random distributions. This will in turn influ-
ence the spherulitic microstructure of the polymer. Block
copolymers are of particular interest as microphase sep-
aration may lead to periodic nanostructures of differing
morphology, with unique mechanical properties as a result.
It should be noted however that in biological systems
the synthesis of a block copolymer is not straightforward.
The kinetics of chain synthesis are fast (see Section 7 and
below), and the intracellular and/or intra-granular concen-
trations of free acids are not necessarily the same as in the
bulk reactor. Likewise, the enzyme packages present within
the bacterial cells will differ depending on bacterial type,
culture/feeding history and so on. The high degree of diver-
sity found in biological systems suggests a widely varying
population of polymer chains may be synthesized within
the bacteria [288]. The final product could well be a com-
plex mixture of random copolymers and/or homopolymers
blended with copolymeric materials of a blocky character,
and care is needed in characterizing the final products.
Core–shell structures of PHA have been prepared by
a number of authors. Curley et al. [289] showed that
a core–shell structure resulted when P. oleovorans was
grown on a mixture of nonanoic acid and 5-phenylvaleric
414 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
acid, with poly(3-hydroxynonanoate) in the core and
poly(3-hydroxy-5-phenylvalerate) around this core. From
SDS-PAGE it was shown that both polymers resulted from
the same enzyme system. A random copolymer could
not be obtained. Kelley and Srienc [290] also produced
core–shell polymer structures in Ralstonia eutropha by
growing the bacteria on fructose and limited quantities
(1 g L−1 ) of valeric acid; RuO4 staining was used to see the
structures (with P(3HB-co-3HV) copolymer taking up more
stain than P(3HB), being less crystalline). Yoon and Choi
[291] studied intracellular degradation rates for PHA inclu-
sions in Hydrogenophaga pseudoflava cells, where a range
of structures varying from homopolymer blends to copoly-
mers to single homopolymer were formed. It was noted
that a blend type structure (or core–shell or similar) was
formed with sequential feeding of glucose for 22 h followed
by butyrolactone (for P(3HB) and P(4HB) blocks), or by
feeding glucose first and then valerolactone (for P(3HB) and
P(3HV) blocks). Pederson and Srienc [292] also produced
a core–shell granule structure, with either P(3HB-co-3HV)
core and P(3HB) shell (produced using either a 4 h or 7 h
valerate feed followed by fructose feeding) or P(3HB) core
with P(3HB-co-3HV) shell (produced with either a 5 h or
10 h feed of fructose followed by valerate feeding). Stain-
ing demonstrated that two phase structures were achieved.
This work was backed by model predictions. It was noted
by these authors that block copolymer structures could be
achieved if feeding pulses are on the timescale of individual
polymer chain synthesis, which was estimated to be 1 h or
less.
Madden et al. [293] examined the kinetics of PHA
synthesis: PHA synthase was said to catalyze 6.7 propa-
gation reactions per second. As a result they calculated
that it should take 0.3 h (18 min) to complete a 0.7 × 106 Da
chain. Therefore to get a block copolymer, it was proposed
that reagents should be changed every 5 min. Mantzaris
et al. [294] developed kinetic models and a theoretical
underpinning of synthetic conditions for the formation of
block copolymers. In terms of P(3HB-co-3HV)-P(3HB) and
P(3HB-co-3HV)-P(3HB)-P(3HB-co-3HV) di- and tri-block
copolymer formation, the optimum conditions were pre-
dicted to require 39 carbon source switches (to give 50%
of total polymer as a block copolymer). This translated to
almost 30 min for each P(3HB-co-3HV) producing stage and
a little more than 1 h for each P(3HB) producing stage. Kel-
ley and Srienc [295] expanded the theoretical concept that
either core–shell or block copolymeric structures for PHA
could be produced by controlling timing of feed sequences.
Chanprateep et al. [251] synthesized possible block
copolymers of HV and HB (with 7 and 10 mol% HV) by
switching ethanol and pentanol as carbon sources; samples
were fractionated using chloroform/heptane. The kinetics
of nonisothermal crystallization was studied using DSC and
analyzed using Ozawa’s equation via three different meth-
ods; the crystallization rate of the nominal block polymers
was found to be faster than that of the random copoly-
mers. The n values from the Avrami equation for P(3HB)
homopolymer and for the block copolymer were both 1,
in contrast to the random copolymer (n ∼ 3). A model pre-
dictive controller (MPC) coupled with a metabolic reaction
model controller was used for controlling ethanol and
n-pentanol concentrations. In a subsequent study [296],
both the carbon to nitrogen ratio and the feeding strategy
were controlled. The crystallization mechanisms for the as-
obtained “blocky” P(3HB-co-3HV) were similar to that for
P(3HB), explained as being due to the P(3HB) crystal com-
ponent dominating. The rate of non-isothermal crystalliza-
tion of this blocky P(3HB-co-3HV) was found to be faster
than that of random P(3HB-co-3HV) but slower than that of
P(3HB). Both normal and blocky P(3HB-co-3HV) materials
were fractionated; the former had D values (by NMR) typi-
cal of random copolymers (see Section 9.1) while the three
fractions from the latter had values compatible with P(3HB-
co-3HV)s rich in long blocks of 3HB units. The nominal block
copolymer was shown to have slower biodegradation rates.
Pederson, McChalicher and Srienc [297] in a subsequent
study used off-gas mass spectroscopy to control valerate
additions for PHA block copolymer synthesis using valeric
acid and fructose addition rates based on oxygen uptake
rate and carbon dioxide evolution rate. Alternating feeding
times of 8, 10, 14 and 32 min gave in theory 21/22, 26/23,
36/24 and 48/10% di/triblocks respectively. The production
of triblock copolymer with equal HB/HV content was tar-
geted using 12.5 min of mixed valerate and fructose feed
alternating with 25 min of fructose. The block copolymer
approach was said to be complicated by the fact that syn-
thesis does not occur simultaneously on all polymer chains
because of random chain initiation and termination events.
About 30% of the total polymer sample revealed melting
properties suggestive of block copolymers.
Mumtaz et al. [121] used TEM to examine P(3HB-co-
3HV) inclusions in wild type Comamonas sp. EB172, a local
isolate, grown on mixed organic acids obtained from palm
oil mill wastewater. Core–shell structures were evident
through RuO4 staining; the authors proposed the forma-
tion of a block copolymeric material, although the DSC data
was also consistent with a blend material and no NMR data
was presented.
Li et al. [298] used the PHA synthesis genes phaPCJAC
cloned from Aeromonas caviae transformed into Pseu-
domonas putida KTOY06C (phaPCJA,c ) to produce a
material consisting of P(3HB) as one theoretical block
(nominally at 70 mol% of the total polymer) and the ran-
dom copolymer P(3HV-co-3HHp) as another block (at
30 mol%), through sequential substrate feeding. Manu-
factured blends as well as random copolymers and the
homopolymers were compared with the blocky material
using NMR, DSC, and cone plate rheology, and the latter
was found to have distinctly different characteristics.
McChalicher and Srienc [288] also prepared block
copolymer (P(3HB)-block-P(3HB-co-3HV)) using C. necator
and compared it with random copolymers of P(3HB-
co-3HV). 30 min of fructose/valeric acid feeding was
alternated with 1 h of fructose-only feed to give 50/50
block copolymers. Likewise, Pereira et al. [299] reported
the probable formation of a P(3HB)-block-P(3HB-co-3HV)
when R. eutropha was fed fructose in the first and propionic
acid in the second stage using “alternating feeding” – the
timing and sequence of this was not described. The prod-
uct, even after fractionation, did not appear to be a blend
of P(3HB) homopolymer and P(3HB-co-3HV) copolymer
based on the NMR, crystallinity and DSC data (see Section
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
9.2.1). Finally Ivanova et al. [81] explored the use of alter-
nating feeding of acetate and propionate to an enriched
mixed culture, and found that 13 C NMR data indicative of
blocky copolymers resulted, although it was not possible
to clearly determine if blends of random copolymers were
formed instead.
In all of the work above, the structural and composi-
tional similarity between the P(3HB) and P(3HB-co-3HV)
blocks means that many quantitative assessment tech-
niques (such as X-ray scattering) are complicated to apply
[288]. Qualitative assessment techniques using a com-
bination of fractionation, calorimetry, rheological, and
nuclear magnetic resonance techniques have been the
tools used to date [297]. Additional assessment tech-
niques may be of benefit. High-resolution solid state NMR
(via 13 C magic angle spinning with dipolar decoupling
and cross-polarization) is a useful technique for pro-
bing the molecular structure, conformation and dynamics
of polymer chains at the interphase between phases in
multicomponent polymer systems, and has been used to
distinguish PE–PP copolymer blends from block copoly-
mers [25,26]. MALDI-TOF analysis of partially hydrolyzed
PHA could also be used to distinguish blocky from random
copolymers [407]. In addition, the molar Kerr constant, as
determined in an electrical birefringence experiment, is a
macroscopic property of the polymer chain that is sensi-
tive to local polymer microstructures, including tacticity
and comonomer sequences [408]. Infrared and Raman
spectroscopy also has the potential to provide useful infor-
mation, given that a characteristic distinguishing feature
of blocky copolymers is microphase separation which can
lead to more ordered molecular environments. These and
other characterization techniques could provide insight
into these as-produced materials.
The effect of these alternative polymerization strate-
gies on the mechanical and rheological properties of PHA
is discussed in Section 11.3.
9. Comonomer sequence compositional
distribution
Given that blends of P(3HB) and P(3HB-co-3HV) can
form different crystalline phases depending on the HV
content of the copolymer, it is important to know the
actual compositional distribution of microbial PHA. To date
the discussion of PHA polymer properties has been based
on the presumption that the bacterial copolymer being
analyzed is a random or tailored copolymer of narrow poly-
dispersity, which can be treated as a pure sample with
well-defined and reproducible properties. However, this
assumption is not necessarily valid.
9.1. Characterization of comonomer compositional
distribution
Many as-produced P(3HB)-based copolymers have
been fractionated using a solvent/nonsolvent fractionation
technique to give a series of fractions with narrow composi-
tional distributions (as judged by 13 C NMR) [197,300–302].
Comonomer composition appears to be the determining
characteristic by which fractions are separated, although
415
Fig. 8. Effect of HV content on solvent fractionation of P(3HB-co-3HV)
using chloroform/hexane.
Reprinted (adapted) from Wang et al. [201], Copyright (2001), with per-
mission from American Chemical Society.
at the same composition fractionation based on molecu-
lar weight can occur. The driving force for the separation
of differing comonomer compositions in mixed solvents is
believed to be the different polarities of the side chains,
with molecular weight being a secondary factor; however,
there is some inconsistency in this mechanism in that as
the mol% 3HV increases above 70%, the solubility in hex-
ane decreases (Fig. 8). It has been proposed that this is due
to the increased relative crystallizability of the materials of
higher 3HV content.
The chemical compositional distribution (CCD) of
P(3HB-co-3HV) is usually determined by 13 C NMR spec-
troscopy. Diad and triad sequence analysis is used to assess
the extent of deviation of the copolymer composition from
the statistically random (Bernoullian) compositional distri-
bution. The parameter D is given by:
FBB FVV
D= (3)
FBV FVB
where FBB , FVV , FBV , and FVB are the fractions of the BB,
VV, BV and VB diad sequences, respectively [303]. The
D value for statistically random copolymers is 1.0. Non-
random copolymers will have D values greater or less
than 1, with “blocky” and alternating copolymers having
D and R values 1 and very close to 0, respectively. How-
ever, while polymers with a high D value could be true
block copolymers, they may also be a mixture of copoly-
mers with different compositions or of P(3HB) and P(3HV)
homopolymers. Therefore D values much higher than 1
clearly indicate bimodal (or multimodal) chemical compo-
sitional distribution but give little information on the width
of CCD. The parameter R, which is an indicator of the degree
of randomness obtained from analysis of the triad distri-
bution, is more sensitive to the broadness of the chemical
compositional distribution. This parameter has a value of
1 for a completely random distribution of HB and HV units
in the copolymer chain and 0 for a diblock copolymer.
Yoshie and Inoue [302] noted that even if the same
author(s) synthesize two samples from the same carbon
source by the same organism under the same fermen-
tation conditions, one sample may be a homogeneous
416 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
random copolymer and another may be a mixture of ran-
dom copolymers. They concluded that the origin of the
complex CCD in bacterial copolyesters is unclear, although
Tanadchangsaeng et al. [304] proposed that varying CCD
can result from a range of factors such as changing bacterial
strains, carbon substrates, pH in the fermentation medium,
physiological state and so on during the cultivation. Even
materials with D values close to 1 (D = 0.994–1.5) could
be fractionated into several fractions [301]. It was shown
that samples with very large D values bear multimodal or
bimodal CCD, while samples with D ≈ 1.5 had a broad CCD.
As an additional technique, some authors use a statistical
approach looking at the relative fit of the data to a Bernoul-
lian model, a first-order Markovian model, a Newton model
and a mixture Markovian model [305–307].
Several papers have also reported on other meth-
ods for determining the comonomer composition of
P(3HB-co-3HV), including FFT-IR spectroscopy, electro-
spray ionization multistage mass spectrometry (ESI-MSn )
[308] and fast atom bombardment mass spectroscopy
(FAB-MS) [309], etc. Much of the early research (prior to
1999) characterizing CCD by means of D values has been
summarized by Yoshie and Inoue [302]. It is beyond the
scope of this review to capture all of the subsequent NMR
studies; the following sections therefore firstly highlight
more recent studies of particular relevance where CCD
only is characterized, and then summarize the results from
actual fractionation studies on as-produced PHA.
Some recent analyses of sequence distributions in
P(3HB-co-3HV) and related copolymers are of note. Žagar
et al. [310] analyzed the monomer distribution in bacterial
P(3HB-co-3HV) containing 12.9 mol% HV (from Ralstonia
eutropha) and 12.0 and 27.1 mol% HV (produced using
an osmophilic bacterium). Both 13 C NMR (diad and triad
analysis) and multistage electrospray ionization mass
spectrometry (ESI-MS) were used to characterize the com-
positional distribution, with the results being in good
agreement for all samples. The 12.0 mol% HV sample had
a much higher molecular weight and narrower polydisper-
sity than the other samples, with a sharp single melting
peak (Tm = 130 ◦ C) and narrow CCD. The other samples
were mixtures of random copolymers of widely different
comonomer-unit composition. A 1:1 mixture of two P(3HB-
co-3HV) copolyesters (containing 12.0 and 27.1 mol% HV
respectively) was also prepared and had D and R values
that were between those of the constituent materials, with
a single melting peak.
Dai et al. [73] used glycogen accumulating orga-
nisms (GAOs), found in mixed culture anaerobic–aerobic
wastewater treatment processes, to produce PHA polymers
consisting of 3HB, 3HV, 3H2MV and 3H2MB monomer units
with a non-HB fraction of up to 37%. The D value was found
to be between 2.4 and 4.7 for different runs and the prod-
uct was more likely to be a mixture of random copolymers
(from the anaerobic production period) with an additional
segment of pure HB blocks or P(3HB) homopolymer (from
the aerobic production stage); it is possible that a poly-
mer of blocky character was formed due to the sequential
uptake of reagents. In a related study [74], similar PHA
copolymers were prepared by manipulating the ratio of
propionate to acetate: less 3H2MV and 3H2MB units were
present as acetate in the feed increased. When propionate
alone was used as a feed, only a random copolymer was
formed, while other PHAs were found to be a mixture of
random copolymers.
Lau et al. [311] prepared P(3HB-co-19 mol% 3H4MV)
using a Burkholderia sp. (wild type) fed with fructose,
crude palm kernel oil, and sodium valerate along with 4-
methylvaleric acid. A D value of 4 was obtained for this
polymer, indicating it was most likely a blend of copoly-
mers.
9.2. Solvent fractionation
There have been many studies looking at in particular
the chemical but also the molecular weight fractionation of
pure culture materials, and one examining mixed culture
PHA. The thermal property and crystallinity data from the
finely fractionated materials is more consistent than for as-
produced materials and should by preference be used as a
reference for comparison (see Fig. 5 and references given
below).
9.2.1. Solvent fractionation of P(3HB-co-3HV)
A summary of results for the fractionation of as-
produced P(3HB-co-3HV) across a broad range of initial
compositions is given in Table 1. In most cases, a very broad
CCD was evidenced, with the as-produced material being
shown to be a blend of random copolymers of differing 3HV
content. It was consistently found that the polymers were
fractionated firstly according to 3HV unit content; how-
ever, when the 3HV-unit contents of fractions were similar,
then they were fractionated according to differences in
molecular weight. For example, Don et al. [312] produced
P(3HB-co-3HV) of 10.8 mol% HV content using Haloferax
mediterranei. Fractionation using chloroform/acetone pro-
duced two compositionally different co-polymers. The
major fraction (at 93.4 wt%, with 10.7 mol% of 3HV) had a
high molecular weight of 5.7 × 105 Da. The second fraction
(ca. 12.3 mol% HV) had a much lower molecular weight of
0.8 × 105 Da.
Four commercial samples were also shown to have
broad compositional distribution in the bulk [313]
(Table 1); for three of these materials, the melting temper-
ature and spherulitic growth rate corresponded to that of
the average composition in spite of their complex composi-
tion distribution. However, one as received P(3HB-co-3HV)
with 21.8 mol% 3HV had a higher melting temperature and
faster growth rate than might be expected from its overall
composition, which was attributed to the more rapid crys-
tallization of the component chains of relatively low 3HV
content.
PHA produced using mixed cultures has also been frac-
tionated. Inoue et al. [314] used an enriched mixed culture
sludge (prepared using a continuous anaerobic/aerobic cul-
ture process) to accumulate PHA using propionate as the
sole carbon source. The resulting as-produced polymer
was composed of four types of monomer units (3HB, 3HV,
3H2MB and 3H2MV) and could be fractionated into at least
four fractions of random copolymers of very different com-
position and molecular weights using water/acetone mixed
solvent.
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442 417

Table 1
Solvent fractionation results for as-produced bacterial P(3HB-co-3HV).

Organism 3HV content of initial # of Range of 3HV Solvent system Character of resulting Ref.
material fractions content in fractions fractions
(mol% 3HV) (mol% 3HV)

46 4 23–72 Random copolymers by

Ralstonia eutropha Acetone/water [193]
64 6 45–85 NMR

24 9 21–31
Random copolymers by
Ralstonia eutropha 36 7 19–55 Acetone/water [194]
NMR
61 8 28–78

Burkholderia cepacia D1 36 5 10–81 Acetone/water Random copolymers by [203]

NMR

45 9 26–60
49 7 27–68
Assumed random
Ralstonia eutropha 70 5 30–74 Chloroform/n-heptane [201]
copolymers by DSC
80 12 54–94
96 8 74–98

45 10 26–60 Assumed random

Ralstonia eutropha Chloroform/n-heptane [202]
47 8 26–56 copolymers by DSC

Sourced from ICI 6.5 3 6.1–15.3

Sourced from ICI 15.3 5 10.9–25.3 Assumed random
Chloroform/n-heptane [313]
Sourced from Aldrich 19.4 5 15.7–24.9 copolymers by DSC
Sourced from Aldrich 21.8 7 10.2–34.4

The fractionation of nominally blocky or core–shell
copolymers of P(3HB-co-3HV) has also been investigated.
Madden et al. [293] produced mixtures of homopolymer
P(3HB) and random copolymer P(3HB-co-3HV) with Ral-
stonia eutropha by alternating long feeding periods of 5 h or
more of glucose and propionic acid, a result that was sup-
ported by solvent fractionation studies. This material had
thermal properties that were markedly different from ran-
dom copolymers of similar monomer content yet exhibited
a single glass transition and a single melting peak that was
significantly higher than expected for a random copolymer.
The P(3HB-co-3HV) fraction that was produced had prop-
erties that were identical to those of the normal random
copolymer of the same HV content.
Pederson, McChalicher and Srienc (in their key paper on
block copolymer formation [297]) fractionated the blocky
P(3HB-co-3HV) material produced using a co-feeding time
of 14 min, using a chloroform/heptane fractionation sys-
tem. Analysis of the four resulting fractions showed a very
broad compositional distribution (from 0% to 44% 3HV).
In addition, the third fraction (which was 30% of the total
weight) had a D value of 9.1, compared to only 3.1 for F4
and 1.9 for F2, indicating a potentially blocky character for
this fraction. Multiple melting peaks were present in the
DSC scan of this fraction indicating that complex blends of
blocky and random copolymers were likely present. Žagar
and Kržan [303] also fractionated two P(3HB)-block-P(3HB-
co-3HV) polymers according to molecular weight using
preparative size exclusion chromatography. These materi-
als were formed using alternating feeding in pure cultures.
For the lower HV content (12.9 mol% HV) material there
was almost no difference in D or R parameters for the dif-
ferent molecular weight fractions, while for the 27 mol% HV
material there was a slight increase in D value (from 6.89
to 8.43) as the molecular weight decreased; however the
R value remained relatively unchanged. Multiple melting
peaks could be seen for these complex mixtures.
9.2.2. Solvent fractionation of other PHAs
A number of studies have looked at the fractionation
of other PHA copolymers (Table 2). In a series of investiga-
tions, as-produced P(3HB-co-3HP) microbially synthesized
by Alcaligenes latus [217,315–320] was found to have a
broad CCD consisting of blends of copolymers. The HP con-
tent of the fractions decreased with increasing n-heptane
content in the chloroform/n-heptane solvent mix. A min-
imum melting temperature was found at 60 mol% HP
content for the fractions produced; only the P(3HB) type
of lattice structure was found at low 3HP unit content,
with the 3HP lattice found at high 3HP contents and an
amorphous polymer at intermediate 3HP contents. The
presence of the minor 3HP comonomer units significantly
suppressed the crystallization of the 3HB-rich copolymers
[317]. It was found that an average block length of more
than two HB units was needed to form P(3HB)-type crys-
tallites [217]. In a further extension [320], the effect of
pH was evaluated (Table 2). The more basic the fermen-
tation medium, the less 3HP units were introduced into
the copolyester, with the narrowest CCD being obtained
at pH 7.0. The distribution became broader with increas-
ing fermentation time. Manufactured blends of P(3HB) and
P(3HB-co-3HP) fractions have also been prepared: it was
found that at 11.3 and 14.9 mol% 3HP, blends with P(3HB)
were miscible in the melt, with crystallization of the P(3HB)
component dominating and intermolecular interactions
between P(3HB) and P(3HB-co-3HP)s found to be present
[318,319].
The solvent/non-solvent fractionation of P(3HB-co-
3HHx) also gave evidence of a broad CCD for most of these
copolymers as produced (Table 2). Again, composition was
the primary driver for fractionation, followed by molecu-
lar weight, and the thermal and crystallization behavior as
well as the crystalline morphology was strongly affected by
the compositional distribution [321–324]. Asrar et al. [324]
found that P(3HB-co-3HHx) copolymers produced using
 418
Table 2
Solvent fractionation results for as-produced bacterial 3HB-containing copolymers (PHAs) with comonomers other than 3HV (3HP, 3HHx, 4HB, 3H4MB).

Copolymer Organism Comonomer # of fractions Range of comonomer Solvent system Notes Ref.
content of initial content of fractions
material (mol% (mol% comonomer)
comonomer)

P(3HB-co-3HP) 26.4 4 14.8–54.2

Non-random
Alcaligenes latus 30.6 11 8.5–60.5 Chloroform/n-heptane [315]
copolymers or blends
62 6 41.3–85.3
36.5 10 13.2–73.6 Blend of random
Alcaligenes latus Chloroform/n-heptane [317]
68.1 12 31.4–95.9 copolymers
Alcaligenes latus 19.4 9 11.3–85.2 Chloroform/n-heptane [319]
A. latus pH 6 33.4 5 20.5–62.3

B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442

A. latus pH 6.5 23 7 15.1–31.4
A. latus pH 7 16.1 5 11.6–17.7 Chloroform/n-heptane [320]
A. latus pH 7.5 20.9 6 13.3–27.1
A. latus pH 8 13.2 4 11.7–37.5

P(3HB-co-3HHx) 7.5 12 5.6–12.9

11.2 21 8.4–18.1 Note refractionation of
Ralstonia eutropha Chloroform/n-heptane [321]
12.3 18 8.1–20.2 some fractions
13.7 21 8.1–21.5
7.4 6 4.4–10.6
Ralstonia eutropha 18 15 7.4–35.6 Chloroform/n-heptane Some fractions still [322]
a blend by DSC
22 13 14.8–41.9 Some fractions still
a blend by DSC
Aeromonas hydrophila 13.8 9 9.2–17.1
Ralstonia eutropha 35.2 2 29.8 and 69.8 Fractionated at
Chloroform/n-heptane [323]
70 ◦ C
Ralstonia eutropha 24.8 2 19.5 and 80.7 Fractionated at
70 ◦ C
Ralstonia eutropha 54 4 50.5–66.3
Aeromonas hydrophila 25.7 2 6.9 and 33.6 Chloroform/n-hexane Blend of random [324]
copolymers

P(3HB-co-4HB) Ralstonia eutropha 24 2 7 mol% and 86 mol% Acetone Blend of random [325]
copolymers
Ralstonia eutropha 28 3 1, 28 and 82 mol% Chloroform/acetone Blend of random [326]
then methanol/acetone copolymers
16.7 7 6.3–46.4
Ralstonia eutropha 50.9 10 7.1–95 Chloroform/n-hexane [327]
69.8 9 11–99.4

P(3HB-co-3H4MV) Ralstonia eutropha 7 4 2–39 Blend of random

(modified) Chloroform/n-hexane copolymers [304]
11 5 3–28 Blend of random
copolymers
35 3 20–47 Blend of random
copolymers
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
Aeromonas hydrophila had a Bernoullian (random) compo-
sitional distribution with 3HHx content up to 9.5 mol%. By
contrast, copolymers produced using a recombinant strain
of Pseudomonas putida GPp104 had >9.5 mol% 3HHx con-
tent and a bimodal polymer composition and were found
to be blends of random copolymers.
Three studies have evaluated the fractionation of
P(3HB-co-4HB) copolymers. Doi et al. [325] found that
while many of the as-produced materials had D values
close to 1, indicating that they were random copolymers,
others had much higher D values (up to D = 66). These
were fractionated into acetone-soluble (with high 4HB
content, ∼86 mol%) and acetone-insoluble (<7 mol% 4HB)
fractions, which each also had D values close to 1, indicating
the original polymers were bimodal mixtures of random
copolymers that had very different compositions. Shi et al.
[326] used a similar approach but further fractionated the
acetone soluble material with methanol/acetone into solu-
ble and insoluble components. The three fractions resulting
were found to have narrow compositional distribution and
were composed of P(3HB), P(3HB-co-28 mol% 4HB), and
P(3HB-co-82 mol% 4HB), respectively, again indicating the
broad compositional distribution in the as-produced poly-
mer. In a more detailed study, Ishida et al. [327] very
finely fractionated three as-produced P(3HB-co-4HB) sam-
ples using chloroform/n-hexane mixed solvent. Again, they
found that the as-produced material had an extremely
broad compositional distribution. In addition, the 4HB-rich
P(3HB-co-4HB) polymer fractions were immiscible with
the 3HB-rich ones. Thus, as-produced original P(3HB-co-
4HB) samples should be considered as immiscible polymer
blend systems.
As-produced P(3HB-co-3H4MV) was also found to have
a broad CCD (Table 2) [304]. Both the melting tempera-
ture and glass-transition temperature of the copolymers
decreased with an increase in the 3H4MV content. The Tm
dropped from 172 to 99 ◦ C and the enthalpy of fusion fell
from 92 to 9 J g−1 , while Tg decreased with an increase in
3H4MV content. The rates of enzymatic erosion initially
increased markedly with an increase in the 3H4MV content
to 7 mol%, then decreased. The crystallinity of fraction-
ated P(3HB-co-3H4MV) films decreased from 60 to 13% (by
WAXS) as the 3H4MV fraction increased from 0 to 39 mol%.
Crystalline structures were those of P(3HB) and the d spac-
ing remained constant with mol% 3H4MV; 3H4MV was
excluded from the P(3HB) crystal [304].
Valentin et al. [328] prepared 3HB and 3-hydroxy-4-
pentenoic acid (3HPE) containing PHAs using Burkholderia
sp. DSMZ # 9243. From solvent fractionation of the result-
ing polymers it was shown that the resulting material was
a blend of two homopolymers poly(3HB) and poly(3HPE),
rather than a co-polyester with random monomer distri-
bution.
Chen et al. [329] fractionated mcl-PHA containing
high proportions of 3-hydroxydodecanoate (3HDD) and 3-
hydroxytetradecanoate (3HTD), using mixed solvents of
chloroform/ethanol. The first fractions had both higher
3HTD composition and higher molecular weight, although
fractions with relatively low Mn but extremely high 3HTD
content were also precipitated earlier. The samples with
higher 3HTD composition had in general higher Tm and
419
enthalpy of fusion. Two different crystal structures were
identified, which were temperature dependent.
Blends of P(3HB) and mcl-PHA rich PHA were also pro-
duced when Pseudomonas species was fed fructose and
mannose, sodium glutamate, glucose, octanoic acid or oleic
acid [88,139,330,331]; these were fractionated using boil-
ing acetone to give in general an acetone-insoluble P(3HB)
fraction and an acetone-soluble fraction composed of mcl-
PHA.
Yu et al. [332] used chloroform/heptane fraction-
ation to fractionate novel sulfur-containing biopolymers,
poly[(3-hydroxybutyrate)-co-(3-mercaptopropionate)]s
[P(3HB-co-3MP)s]; a sample with 3MP unit content of
16.3 mol% was fractionated into eight fractions with 3MP
unit content ranging from 10.3 to 37.2 mol%. Both the 3MP
and the 3HB crystal units were present in some fractions,
either because the material was naturally blocky in char-
acter or because there were 3MP and 3HB rich phases.
The earlier fractions were highly blocky in character (as
assessed by D value) while others were more random;
however, there was no regular progression in terms of
increase in mol% 3MP units with increasing heptane con-
tent, nor was there a regular pattern in terms of molecular
weight distribution.
In conclusion, solvent fractionation has proved to be an
effective method for production of copolymer PHA with
narrow CCD. Many as-produced copolymers have been
shown to be composed of a broad range of compositions,
so the properties of the as-produced materials are often
different to the component copolymers – it is important to
establish the CCD of any PHA produced using bacterial PHA.
Whichever method is used for solvent fractionation, care
must be taken to ensure that the polymer is not degraded
during the sample preparation and fractionation stages
since it has been shown that dissolution of the polymer
followed by precipitation can cause a severe decrease in
molecular weight [22], particularly for chloroform/alcohol
combinations over time. In addition chloroform often con-
tains ethanol as a stabilizer; refluxing for extended periods
in this can result in significant esterolysis particularly if acid
impurities are present to catalyze the reaction.
10. PHA–PHA blends
Given that many as-produced microbial PHA polymers
consist of blends of copolymers of differing composition,
it is obviously important to understand the effect of such
polymer blend systems on polymer properties.
Chain isomorphism can occur with blends of different
P(3HB-co-3HV) copolymers or with P(3HB)/P(3HB-co-
3HV) blends [14] whereby two or more different
semicrystalline polymers are miscible in the crystalline as
well as the molten state. In such cases, the blend will exhibit
single composition-dependent values of Tg and Tm . How-
ever, this is not the general case for PHA: there have been a
number of studies exploring the effect of deliberate blend-
ing of different random copolymers of PHA on polymer
properties, and it has been shown that there is only a nar-
row window of composition in which chain isomorphism
can occur.
420 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
Fig. 9. Variation of phase structures in PHB/PHB-HV blends.
Reprinted (adapted) from Yoshie and Inoue [337], Copyright (2005), with permission from John Wiley and Sons.
Kamiya et al. [305] prepared three blends of bacterially
synthesized random copolymeric P(3HB-co-3HV) (contain-
ing 23.4, 40.7 and 92.5 mol% 3HV respectively) with P(3HB),
as well as a blend of the 40.7 and 92.5 mol% 3HV copoly-
mers, using solvent casting. Two melting peaks were found
for each sample, except for the mix of P(3HB) and 23.4 mol%
P(3HB-co-3HV) that had a melting point corresponding
to pure P(3HB) homopolymer. Melting peak temperatures
were independent of the relative amount of each polymer,
indicating that there were different phases present. Kuma-
gai et al. [333] prepared P(3HB)/P(3HB-co-76 mol% 3HV)
blends and found that all blends were immiscible based on
Tg data. The rate of enzymatic degradation was found to be
higher for the blends than for the individual components,
perhaps due to the large holes that formed. Barham, Organ
and coworkers [211,240,241,334] also showed that blends
containing above 60–70% P(3HB-co-3HV) (18.4 mol% HV)
with P(3HB) were phase separated in the melt, with two-
phase crystallization occurring. There was some evidence
of partial miscibility between the blend components as
judged by a region of liquid–liquid phase separation, and
there was a kinetic effect associated with the characteris-
tics of the phases. Saito et al. [335] reported the almost
perfect co-crystallization of P(3HB) and P(3HB-co-3HV)
when the HV content was less than 9 mol% 3HV, whereas
phase segregation preceded the crystallization in blends
when the 3HV content was >15 mol%. It was proposed
that the component of slower crystallization is partially
excluded from the crystals (since P(3HB-co-3HV) with a
lower crystallization rate moves away from the growing
front) resulting in a P(3HB)-rich crystalline phase with
thicker amorphous layers. When the difference exceeds
30–40%, the blends become immiscible [336]. The co-
crystallizable blends by contrast had the same properties
as the true copolymer of same overall 3HV content. It was
found that the melting temperatures reflected the blend
type; blends with more than two crystalline phases had two
or more melting peaks, while only one melting peak (at the
temperature expected from the average HV content) was
observed when blends formed complete co-crystals. Yoshie
and Inoue [337] found that PHB/PHBV blends exhibited a
spectrum of properties ranging from complete cocrystal-
lization (with rapid crystallization rates) through partial
segregation with some cocrystallization to complete phase
segregation depending on the HV content of the copolymer
(Fig. 9). If the crystalline phase is rich in the faster-
crystallizing structures, then the melting temperature will
be higher than expected, and at least two melting temper-
atures will be observed in phase separated systems [223].
Other blend mixes have also been explored. Pearce
et al. [223] blended P(3HB-co-3HV) (77 mol% 3HV) and
P(3HV) at a 50/50 ratio; a single melting peak was observed
plus a linear Hoffman–Weeks plot was obtained, implying
that co-crystallization occurred. The temperature of maxi-
mum crystallization rate decreased as mol% HV decreased.
Blends of P(3HV) and P(3HB) were also prepared (at 77%
HV load): individual glass transition temperatures were
obtained showing that the homopolymers phase separated
in the melt. This suggests that, for the copolymer, the co-
crystallization may be partially kinetic in origin. Gassner
and Owen [225] also investigated the blends of P(3HB)
and P(3HV). Their results show that melt-pressed blends
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
contain phase-separated domains in the melt that then
crystallize as P(3HB) and P(3HV) spherulites respectively.
The two melting regions were detected by DSC, and were
almost unaffected by blend composition. A summary of
these outcomes was prepared by Yoshie and Inoue, and is
given in Fig. 9 [337].
The effect of molecular weight has also been explored.
High molecular weight P(3HB) (Mw = 0.35 × 106 Da) and
low molecular weight P(3HB) (Mw = 3900 Da) were blended
with P(3HB-co-3HV) (12 mol% 3HV, Mw = 0.24 × 106 Da).
After isothermal crystallization, a lowering of the Tm of
P(3HB) was observed, indicating miscibility in the liquid
and the solid state. Only one melting peak was observed,
indicating co-crystallization [227]. Saad et al. [338]
prepared solvent cast blends of P(3HB) and oligo[(R,S)-
3-hydroxybutyrate]-diol and found that up to 60 wt%
oligo-HB, the blends behaved as single-phase amorphous
glasses with a composition dependent glass transition, Tg ,
and a depression in the equilibrium melting temperature
of P(3HB), indicating that the components were miscible.
The glass transition of melt-crystallized blends contain-
ing 0–20 wt% oligo-HB were dielectrically investigated and
revealed the existence of a single alpha-relaxation process
for blends, indicating the miscibility between amorphous
fractions of P(3HB) and oligo-HB.
Tacticity also influences compatibility between PHA
components. Pearce et al. [339] prepared blends of bacte-
rial and partially isotactic P(3HB); both solvent cast films
and rapidly co-precipitated powders were used. In the
latter case a single melting point was observed (i.e. a sin-
gle phase material was produced), while slow evaporation
led to two endotherms, one for each phase. Totally atac-
tic P(3HB) was found to be incompatible with bacterial
P(3HB) but did cause a significant drop in the melting point.
Abe et al. [88,89] also prepared binary blends of synthetic
atactic or syndiotactic and bacterial isotactic P(3HB) and
found that the equilibrium Tm decreased with the addi-
tion of synthetic P(3HB) from 191 to 174 ◦ C, suggesting that
the blends are miscible in the melt and in the amorphous
state. Crystallinity (based on XRD) also decreased with an
increase in synthetic P(3HB) and the mechanical proper-
ties changed remarkably. For example, Young’s modulus
drops from 1560 MPa to around 180–340 MPa with the
inclusion of 50% atactic or syndiotactic P(3HB) (of vary-
ing % (R)). Scandola et al. [340] also looked at blends of
synthetic atactic P(3HB) with a natural bacterial P(3HB-co-
3HV) containing 10 mol% of 3HV units. These blends were
also miscible in the melt and solidified to give a spherulitic
morphology. The degree of crystallinity decreased as the
content of atactic P(3HB) increased, and the elongation at
break for a sample containing 50% of atactic P(3HB) was
30-fold that of pure P(3HB-co-3HV). The rate of enzymatic
degradation of the blends was higher than that of P(3HB-
co-3HV) and increased with the atactic P(3HB) content in
the blends, whereas pure atactic P(3HB) did not biodegrade
under these conditions. It was concluded that a partially
crystalline P(3HB) was essential for enzymatic degradation.
It is not just the P(3HB-co-3HV) copolymers that display
phase separation; this phenomenon has also been demon-
strated for a range of other PHA copolymers. Hirota et al.
[341] found that on solvent blending P(3HB-co-3HV) (7,
421
12 and 19 mol% 3HV) and P(3HB-co-3HP) (10 and 14 mol%
3HP), the resulting polymers were generally phase segre-
gated with partial co-crystallization, although some blends
co-crystallized. The copolymer with the fastest crystal-
lization rate forms the lamellae and the other copolymer
is involved into the lamellae. The same was observed
by Yoshie et al. [336] who found that the blends of
P(3HB-co-3HV) and P(3HB-co-3HP) adopt various phase
structures depending on the extent of phase segrega-
tion, from complete co-crystallization to complete phase
segregation. The degree of phase separation was found
to depend on the relative isothermal spherulite growth
rate, G, for the two copolymers in the blend. Feng et al.
[342] prepared blends of finely fractionated P(3HB-co-
3HHx) copolymers, including blends with P(3HB), using
solvent casting. It was found that when the difference in
the 3HHx content between the two different copolymers
was less than 20 mol%, then they were miscible in the
amorphous phase; when the difference was larger than
30 mol%, then they were immiscible. In addition, a con-
tinuous compositional distribution was found to favor a
homogenous amorphous phase, while bimodal composi-
tional distributions lead to a phase-separated morphology.
Zhao et al. [250] also blended PHA and Nodax (13 mol%
HHx); the blends were miscible, and had good mechani-
cal properties, e.g. elongation to break increased from 15%
to 106% as % Nodax went from 40% to 60%. Abe et al. [88]
sought to compatibilize blends of P(3HB)/P(3HHx) through
the use of a synthetic block copolymer of atactic P(3HB)
and P(3HHx). The addition of small amounts of this block
copolymer to blends of the two homopolymers led to a
decrease in the size of the P(3HHx) dispersed domains and
improved their mechanical properties. It was found that the
rate order for enzymatic hydrolysis of these samples was
P(3HB)/P(3HHx) blend > ternary blend > P(3HB) > P(3HHx).
In summary, phase separation is a common character-
istic of blends of PHA copolymers of sufficiently different
composition, so in interpreting the properties of any PHA
polymer, it is important to understand the compositional
make-up and distribution of that polymer.
11. Mechanical properties of PHA
All of the preceding methods for controlling compo-
sitional distribution, crystallinity, microstructure, blend
composition and molecular weight were techniques for
manipulating polymer processing/mechanical properties.
Three core mechanical properties should be considered
for commodity applications [288]: (1) the elongation at
break, which is a measure of toughness or total deformation
before fracture; (2) the Young’s modulus which is a mea-
sure of stiffness; and (3) the ultimate tensile strength which
is a measure of strength prior to the onset of permanent
plastic deformation.
11.1. Properties of pure culture PHA
It is well known that P(3HB) homopolymer and to a
lesser extent the P(3HB-co-3HV) copolymers of low HV
content are stiff and brittle, with poor impact strength,
because of the relatively high crystallinity of the materials
422 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
Fig. 10. Effect of aging time on mechanical properties of P(3HB-co-3HV):
+: tensile modulus; : Izod notched impact strength.
Reprinted from de Koning and Lemstra [381], Copyright (1993), with per-
mission from Elsevier Limited.
and in particular the significant secondary crystallization
that occurs post processing with age (Fig. 10). Based on
the data available (Appendix B) the average properties
for P(3HB) are: Tg ≈ 5 ◦ C (by DSC) or 12 ◦ C (by DMTA);
Tm ≈ 176 ◦ C; tensile modulus 2.9 GPa; tensile strength
37 MPa; maximum elongation to break ∼4%. Addition of
a nucleating agent can increase the elongation to break
slightly [343], typically up to 5%.
This brittleness has been a reported obstacle to the
practical applications of these materials, and significant
research effort has been devoted to manipulating these
mechanical properties. They are known to be dependent
on polymer composition and processing conditions, and in
particular on whether or not the polymers were produced
using solvent casting or melt mixing and casting. It was
noted by Holmes [22], for example, that films of P(3HB) and
its copolymers prepared by solvent casting have a very fine
spherulitic morphology because of low crystallization tem-
perature and resulting high nucleation density, which can
result in higher elongation to break and impact strength.
The same was observed for P(3HB-co-3HHx) copolymers
(Appendix B) where the solvent cast films had much higher
elongation to break than melt pressed films, although
tensile strength and Young’s modulus were similar. In
assessing mechanical properties, therefore, it is important
to note how samples are prepared.
One of the most direct solutions for improving mechan-
ical properties of PHA is to use copolymeric materials,
such as P(3HB-co-3HV) with higher HV contents, or mcl-
PHA copolymer. Compared to P(3HB), P(3HB-co-3HV) has
decreased stiffness and brittleness, increased flexibility
(higher elongation to break), and increased tensile strength
and toughness. These improved properties have been
attributed to the presence of dislocations, crystal strain
and smaller crystallites due to the insertion of the 3-
hydroxyvalerate unit into the P(3HB) lattice [179], with
HV units acting as defects in the HB crystals. This results
in significant depression in the melting point, the heat
of fusion of the HB component crystal and the fold sur-
face free energy [172]. Both Young’s modulus and tensile
strength decrease with an increase in HV content (Fig. 11)
Fig. 11. Effect of HV content (mol%) of P(3HB-co-3HV) on Young’s modu-
lus and tensile strength (see Appendix B for references).
although there is a lot of scatter in the reported data. There
is even more variation in the data for elongation to break
(Appendix B), which is a property that is very depend-
ent on processing and testing conditions. Considering the
information presented in Section 9, it is also likely that
some of the materials analyzed are in fact blends, or of
a blocky nature, rather than being true copolymers; only
data derived from random copolymers of very narrow com-
positional distributions that have been allowed to fully
crystallize provides reliable information on correlations
between mechanical properties and measures of crys-
tallinity, microstructure, blend composition and molecular
weight.
Annealing should improve the polymer properties, with
some suggestion that maximum elongation to break up
to 130% could be found for 12% HV even 3 months after
heat treatment [343]. Noohom et al. [344] also noted
that the effects of annealing on mechanical properties for
melt pressed discs of commercial P(3HB-co-3HV) (Fig. 12)
– Young’s modulus increased significantly where tensile
strength was little affected apart from an increase in vari-
ability. It was noted that there was <20% loss in molecular
weight under these conditions. Bloembergen et al. [345]
created synthetic analogs of P(3HB-co-3HV), in which long
blocks of the copolymer chain had either the (R) or the
(S) configuration, by ring-opening polymerization of mix-
tures of racemic ␤-butyrolactone and ␤-valerolactone with
a stereoselective alumoxane catalyst. It was found that
by blending this synthetic P(3HB-co-3HV) with bacterial
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442 423

P(3HB-co-3HHx)

Foam mold

0.8–1.8

120–125
10–100
5–10

10–20
Kaneka

1.2
Extrusion, injection

1.2–1.8
and fibre
Biocycle

15–25

25–30
20–30
2400-5

1.2
P(3HB-co-3HV)

Extrusion and
Fig. 12. Effect of annealing temperature on mechanical properties of
P(3HB-co-3HV): 䊉: Young’s modulus and : compressive strength.

injection

170–175
Biocycle

2.5–6
2.5–3
10–12

50–60
30–40
1.22
Reprinted from Noohom et al. [344], Copyright (2009), with permission

1000
from IOP Publishing.

P(3HB), the lamellae crystals were disrupted somewhat

P(3HB-co-3HV)

Injection mold
and as a result the brittleness was reduced and the tough-
ness increased.

5–10
ENMAT

1.25
The degree of molecular segmental motion the amor-

1.4

1.4
36

61

147
143
phous portion of PHA can achieve is strongly influenced by
the difference between the Tg and the end-use tempera-

Extrusion and injection

ture. Therefore, even at the same crystallinity, a material
with a lower Tg (i.e. a longer side chain) should be more
flexible [12]. The mechanical properties for the mcl-PHA
copolymers P(3HB-co-3HHx), P(3HB-co-4HB) and P(3HB-
co-3H4MV) have also been characterized (Appendix B) P1002
and as expected these materials are more flexible than
Mirel

1.3

1.9
the P(3HB-co-3HV) copolymers. As for P(3HB-co-3HV),

26
13

35

137
P(3HB) copolymers

the modulus and tensile strength for these copolymers

decrease with an increase in comonomer content, while
Injection mold

elongation to break increases. Again, there is very signifi-

Properties of commercial PHA (data reproduced with permission from Shen et al. [406]).

cant variability in elongation to break. It should be noted

P1001

1.39
Mirel

that the mechanical properties of a number of terpolymers

3.2
28
6

46

148
and above have also been characterized, and that the elon-
gation to break for these materials can often be higher than
for the comparable copolymers containing the equivalent
Injection mold

monomer units [346].

Shimamura et al. [347] examined the dynamic mechan-
Biomer

9–13

24–27
60–70
1.25

6–9
P226

ical properties of P(3HB-co-3HP) but to date the static

35

96

mechanical properties of this copolymer have not been

characterized. Gagnon et al. [247] found that poly(3-
Injection mold

hydroxyoctanoate) (PHO) crystallized very slowly, taking

approximately 7 weeks at 20 ◦ C and 16 weeks at 5 ◦ C to
Biomer

reach a constant value. The thermal history had a large

P(3HB)

10–17
60–70
18–20
1.17
5–7
240

effect on the mechanical properties, with Young’s mod-

17

53

ulus varying from 2.5 to 9 MPa, tensile strength at break

from 6 to 10 MPa, and ultimate elongation from 450% to
VICAT softening temperature (◦ C)

300% depending on the crystallization temperature. The

equilibrium melting point of PHO was approximately 68 ◦ C.
Melting temperature (◦ C)
Melt flow rate (g/10 min)

A summary of the properties of some commercial grades

Flexural strength (MPa)
Flexural modulus (GPa)
Tensile strength (MPa)

Tensile modulus (GPa)

of PHA is also given in Table 3, illustrating the current range

Application grade

of properties available from processed resins [11].

Density (g/cm3 )
Crystallinity (%)

Elongation (%)

11.2. Properties of mixed culture PHA

Table 3

The physicochemical properties of PHA produced using

mixed culture production have not been widely reported
424 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
until recently. While in principle the properties of PHA
obtained from mixed cultures should be a combination
of those obtained from pure cultures, the complexities of
these diverse biological systems mean that, as discussed in
the previous sections, the final properties may be hard to
predict and so it is worth considering separately the data
available to date.
The first reports on characterization of the molecular
weight properties of mixed culture PHA were published
in 1998 [348] and were of PHA synthesized using an
EBPR process. In this case P(3HB-co-3HV) copolymers were
produced with an HV content determined by the feed com-
position. A summary of the studies for which thermal data
have been obtained is given in Appendix A.
As Bengtsson et al. [62] noted, there is often a higher Tm
and lower enthalpy of melting for mixed culture materials
than is found in the literature for PHA from pure culture.
One hypothesis is that micro-block copolymers could be
forming based on the feeding strategies adopted [62]. It
was observed that with pulse feeding, there was often vari-
ability in the feed concentration of the VFAs – for example,
Bengtsson et al. [62] found that in the early stages of each
pulse, propionate was present and PHA rich in 3HV was
produced. At the end of the pulse, propionate had all been
consumed and PHA dominated by 3HB was produced. This
alternating feeding reflects the deliberate strategy adopted
for the formation of block copolymers in PHA [81]. It is
likely that other PHA production systems experience a sim-
ilar dynamic in VFA feeding [54], leading also to similar
types of block or blend type polymers. Dai et al. [44,73,74]
studied the production of PHA copolymers using enriched
GAO cultures and also found there was a blocky character to
the materials produced. It is believed that the P(3HB) gen-
erated during the aerobic period formed as separate P(3HB)
homopolymer “blocks”, which led to a high melting tem-
perature. The small fraction of P(3HB-co-3HV) was likely
present in the form of separate copolymeric chains. It was
also found that when propionate was used as the sole car-
bon source, random copolymers of 3HB, 3HV, 3H3MB and
3H2MV were produced, while co-feeding propionate and
acetate resulted in heterogeneous copolymers with non-
random sequence distributions. This was consistent with
sequential uptake of the carbon source, forming HV- and
HMV-rich copolymers from propionate and HB- and HV-
rich copolymers from acetate. By contrast, Inoue et al. [314]
examined the microstructure of the PHAs produced using
EBPR sludge with propionate feed as the carbon source
using 13 C NMR. Four types of monomer units were iden-
tified (3HB, 3HV, 3H3MB and 3H2MV) and in this case the
data suggests they were statistically random copolymers.
There was no true evidence for four-monomer-component
copolymers.
For blocky materials, the crystal unit will likely have
higher HB content than the bulk. For blends, the melting
temperatures will reflect the faster crystallizing compo-
nents in the blend that will be higher in the HB or HV
comonomer unit than the bulk unless they are completely
miscible blends. Either way, the melting temperature is
likely to be higher than expected from the nominal 3HV
content. The lamellae are also likely to be small and
imperfect, leading to lower heats of fusion than would be
expected. In addition Serafim et al. [41] noted that for a
P(3HB-co-3HV) copolymer (30 mol% HV) obtained under
ADF conditions, the as-produced material had an impurity
of molecular weight of 2000; a 25% increase in the enthalpy
of melting was observed when it was removed, indicating
its presence was possibly disrupting the lamellae forma-
tion.
Patel et al. [349] produced low polydispersity, ultrahigh
molecular weight P(3HB) and P(3HB-co-3HV) at 18 wt% and
30 wt% respectively using a mixed culture under conditions
of extreme nitrogen limitation (such that nitrogen fixa-
tion was necessary). Only 3.3% of the carbon substrate was
recovered as PHA, but the properties as reported were very
comparable with pure culture PHA (Appendices A and B).
Consistent with other literature, the Tg seems to be mainly
determined from the HB content.
The thermal stability of mixed culture P(3HB-co-3HV)
materials was assessed by Carrasco et al. [350] using ther-
mogravimetric analysis. They found that these materials
(from biomass enriched using different procedures, but
both containing 20 mol% HV) had a similar thermal stability
(with decomposition temperatures of 266.7 and 273.3 ◦ C)
to the commercially available P(3HB) and P(3HB-co-3HV)
(10 mol% HV). They also found that an increase in organic
load during the mixed culture accumulation was linked to
an increase in thermal stability.
In the only study to date that has reported on the
mechanical properties of mixed culture PHA, Dobroth
et al. [351] enriched a mixed microbial consortium from
a wastewater treatment plant using crude glycerol as car-
bon feed. The P(3HB) that accumulated was produced from
the methanol fraction. The thermal properties and molec-
ular weight are reported in Appendix A; the molecular
weight was slightly low, and there was some variability
in the thermal properties, which may be in part related to
the molecular weight variability, but in general the data
is comparable with that of pure culture P(3HB). The tensile
strength and Young’s modulus of the solvent cast films after
aging were 14 MPa and 1.8 GPa, respectively, which were
lower than generally reported for P(3HB). This could be
because of the lower molecular weight compared to com-
mercial materials (308 kDa c.f. 437 kDa) or because there
was some minor contaminant or copolymeric component.
Tensile strength at 20 MPa was comparable with pure cul-
ture materials as were the Young’s modulus (2.3 GPa) and
elongation at break (1.3%).
In the studies that have produced the highest PHA
contents to date, Johnson et al. [77,78] found that as HV
content increased from 15 to 39 mol%, Mw decreased from
6.5 to 2.2 × 105 Da. The crystallinity by XRD was 40 ± 5%
(which is the first reference in the mixed culture literature
to analysis of crystallinity using this method, and which
is in line with literature expectations). The crystallization
rate was very slow; immediately after melting there were
no crystals formed. After one day a large number of very
small crystals was observed. The crystallization rate slowed
as 3HV content increased.
In summary, it is important to demonstrate that the
quality of the biopolymers produced by mixed microbial
cultures can consistently meet the standards required for
use in commercially interesting plastic applications. More
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442 425

Table 4
Effect of block copolymer structure on thermomechanical properties of PHA.

Sample 3HB 3HV 3HHp 3HHx Tm (◦ C) Tc (◦ C) Tg (◦ C) Mn (×105 ) Mw /Mn Young’s Tensile Elongation
mol% mol% mol% mol% modulus strength at break
(GPa) (MPa) (%)

P(3HB) homopolymer 100 0 0 0 171.8 41.2 3.1 – – 1.47 18 3

Random copolymer 4.4 46.5 49.1 0 – – −26.4 – – – – –
Random copolymer 72.9 13.1 14 0 – – −7.3 1.27 2.9 0.13 7.0 462
Block copolymer 71.5 10.2 16.5 0 170.6 50.6 −23.6,3.5 0.52 8.7 0.37 7.5 63
Blend 71.5 10.2 16.5 0 166.5 54.3 −20.0,4.4 0.81 4.4 0.26 5.3 24
PHBHHx 88 0 0 12 – – – – – 0.28 7.0 400
P(3HV) homopolymer 0 100 0 0 106.2 57.9 −15.7 – – 0.39 6.6 3.5

Data abstracted with permission from Li et al. [298].

fundamental studies on thermomechanical properties of
these materials are needed, and with these one would hope
to gain deeper appreciation of the importance of controlled
or uncontrolled compositional distribution variation on
microstructure and relevant mechanical properties.
11.3. Properties of “block” copolymeric PHA
A number of studies have explored the effect of nomi-
nally blocky coplymeric compositions on mechanical and
rheological properties of PHA. Pederson et al. [297] in their
study of HB/HV di- and triblocks used solvent fraction-
ation to separate homopolymer and random copolymer
from the 30% fraction with block copolymer character-
istics that was present. The resulting blocky materials
showed significantly increased elongation to break – for
example, the polymer formed using an 8 min cycle had
250% elongation to break while having the largest Young’s
modulus and tensile strength of the materials produced.
Additional mesophase transitions were observed during
rheological testing, which are only found in block copoly-
mers, and secondary transitions in the storage modulus
(not found in random copolymers) were also observed.
When McChalicher et al. [288] prepared block copoly-
mer (P(3HB)-block-P(3HB-co-3HV)) it was found that the
random P(3HB-co-3HV) copolymers had less than 20%
elongation at break within a few days after annealing. How-
ever, the block copolymer films retained higher than 100%
elongation at break a full 3 months after annealing. It was
also noted that “because block copolymers covalently link
polymers that would otherwise form thermodynamically
separate phases, the rates and degrees of crystallization of
the block copolymers are less than the random copolymer
samples”.
Finally Li et al. [298] used ethanol fractionation to
remove random copolymer and retain the block copolymer
in a mixture of materials comprising a blocky copolymeric
component containing 70% P(3HB) as one block and the
random copolymer P(3HV-co-3HHp) as another block. A
comparison was made with the random copolymer of the
same composition, as well as the homopolymer of P(3HB)
and the random copolymer of 3HV and 3HHp; the blend of
P(3HB), P(3HV) and P(3HHp) was also prepared to provide
a composition similar to that of the block copolymer.
The 13 C NMR signal of the carbonyl region was used to
distinguish the block copolymer from the blend, as was
rheological analysis, which showed a rapid drop in the
storage modulus with cooling. The properties are summa-
rized in Table 4. Young’s modulus was significantly higher
than for the random or blend copolymers of equivalent
composition, and tensile strength was higher than for the
blend, although elongation to break is significantly lower
than for the random material.
Overall, while materials obtained as block copolymeric
compounds do exhibit unusual properties, there is incon-
sistency in results to date and some detailed structure-
property relationships that correlate comonomer sequence
distributions with end-use properties need to be developed
and detailed characterization undertaken.
11.4. Physical techniques for the modification of PHA
mechanical properties
One common method for introducing modifications to
mechanical properties is to induce strain into the struc-
ture through stretching or melt spinning of PHA fibers
[352]. Miller and Williams [353] used monofilament fibers
of P(3HB) and P(3HB-co-3HV) (from ICI) for implantation
studies in rats. The mechanical properties of the origi-
nal materials were significantly improved as fibers (tensile
strength of P(3HB) of ∼226 MPa at 60 ◦ C for example) –
there was little change in these properties with time in
vivo. Schmack and coworkers [354] obtained P(3HB) fibers
with tensile strengths of 330 MPa and a tensile modulus of
7.7 GPa using high-speed melt spinning and drawing fol-
lowed by annealing (with a draw ratio of 6.9). Both the ␣-
and ␤-form crystals were present (while at a draw ratio of
4, only the ␣-form was found). Iwata et al. [352,355] pro-
duced high strength ultrahigh molecular weight (UHMW)
P(3HB) films with a strength of 287 MPa or ∼400 MPa using
cold or hot two-step drawing respectively; these proper-
ties were stable for 4 months, indicating that secondary
crystallization did not occur for these materials. Yamane
and coworkers [180] obtained P(3HB) fibers with a ten-
sile strength of 310 MPa by a two-step annealing process
of melt-spun fibers; in an extension of this work [185] they
obtained P(3HB) fibers with a tensile strength of 416 MPa
by a combination of cold drawing and two-step draw-
ing methods followed by annealing under tension. Strong
fibers with high tensile strength were produced by Iwata
et al. [355,356] using UHMW P(3HB), by quenching in
ice-water followed by two-step drawing. Tensile strength
increased from 38 MPa to 121 MPa after cold drawing,
and then reached 1.32 GPa after total draw-down reached
426 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
60-times, with an elongation to break of 35%. Using micro-
beam X-ray diffraction, it was shown that the central core
of the fiber contained both the ␣- and ␤-crystal forms,
with an ␣-form outer sheath. It is believed that this is due
to the rotation of the ␣-crystal form lamellar crystals in
the outer sheath during the drawing process, which pre-
vents the molecular chains between the lamellar regions
from elongating sufficiently. A similar process was adopted
for commercial P(3HB) fibers (Mw = 7 × 105 ); a four-times
drawn fiber crystallized for 72 h had a tensile strength of
740 MPa, elongation to break of 26% and Young’s modulus
of 10.7 GPa [355]. In this drawing method, small crys-
tal nuclei grow initially during the isothermal nucleation
process. The molecular chains between the small crystal
nuclei then act as entanglement points that are oriented
by stretching.
Tanaka et al. [357] stretched P(3HB) fibers after isother-
mal crystallization near the glass transition temperature.
Increasing the time for crystallization resulted in a decrease
in the maximum draw ratio. However, if the fibers were
allowed to crystallize for 24 h then the tensile strength
increased markedly. They also used 8% HV P(3HB-co-3HV)
– in a similar way, the ten times drawn fiber had a tensile
strength of 90 MPa. However, ten times drawn P(3HB-co-
3HV) after an isothermal nucleation treatment had a tensile
strength of 1.06 GPa, elongation to break of 40% and Young’s
modulus of 8.0 GPa. In contrast to the P(3HB) fibers, those
made from P(3HB-co-3HV) did not exhibit a core/shell
arrangement. Instead the ␤-crystal form was uniformly dis-
tributed throughout the fiber along with the ␣-form, and
many voids were seen in the cross-sectional analysis.
The spherulites of P(3HB) and P(3HB-co-3HV) of low
3HV content usually contain cracks, which cause brit-
tleness. Radial cracks [358–360] occur even at room
temperature; circumferential cracks appear if films are
crystallized at higher temperatures and then cooled. It has
been suggested that this is due to differences in the thermal
expansion coefficients along the radius and circumference,
which causes large internal stresses. It is possible to make
spherulites without cracks either by using a copolymer
of more than 12 mol% 3HV content, or by increasing the
nucleation density until the spherulites are small enough
to avoid cracking (i.e. <20 ␮m) [300]. Cracks can also be
sealed by rolling the films [300].
11.5. Effect of additives
The use of additives is also a common approach for prop-
erty modification, and a thorough review is beyond the
scope of this report; the following is a brief summary of
some of the approaches that can be adopted. It should be
noted that additives can also be grouped according to their
environmental performance, e.g. traditional additives that
have no impact on health or the environment and do not
limit biodegradability; “renewable” additives derived from
natural sources, which are not necessarily biodegradable;
and additives that are both renewable and biodegradable.
Nucleating agents accelerate the rate of crystallization
and can help limit secondary crystallization. Five types
of nucleation processes have been identified for P(3HB)
[155]. They are: homogeneous nucleation, which happens
when the completely melted polymer is cooled from the
melt; self-seeding, which is nucleation by unmelted crys-
tallites when the polymer is not completely melted (the
maximum temperature is kept to less than 15 ◦ C above the
melting temperature); enhanced self-nucleation, which is
like self-seeding in that when the maximum tempera-
ture is limited to only 1 ◦ C above the melting temperature,
and the original spherulites simply reappear on cooling;
nucleation by impurities such as talc which are coated
with bound crystallites having a higher-than-usual melt-
ing point; and finally nucleation through the addition of
nucleating agents which can initiate nucleation from their
surface. The nucleation rate of PHA copolymer is extremely
low so in general nucleating agents need to be added to
facilitate processing. Spherulites of P(3HB-co-3HV) con-
taining from 0 to 25 mol% 3HV are formed only after
melting above 184 ◦ C. This suggests that seeding of the
copolymer spherulites by P(3HB) nuclei is occurring. There
are a large number of nucleating agents that have been
tried with PHA materials such as boron nitride [255,361],
saccharin [361], lignin [362], terbium oxide (Tb2 O3 ), lan-
thanum oxide (La2 O3 ) [255], ␣-cyclodextrin [363], orotic
acid [364], and phthalimide [365]. Clarifying agents may
also be required to prevent clouding.
The use of impact modifiers, whereby a rubbery additive
such as hyperbranched polymers, chlorinated polyethyl-
ene, polychloroprene and poly(butyl acrylate) is added to
improve toughness is also common; however, this can be
a result of a decrease in yield stress rather than a genuine
increase in the material’s toughness as judged by the crit-
ical stress intensity factor [22]. Plasticizers can also have
a dramatic effect on Tg with concomitant improvements
in ductility and crystallization kinetics at the expense of
lowering softening points [366]. Interactions with many
blends are dominated by the tendency for P(3HB) to self-
crystallize with exclusion of the additive to the crystalline
phase. In general, the more polar the plasticizer, the greater
its impact on the polymer and its compounds, since plas-
ticizers work by interacting with the polymer chains,
increasing the free volume and thus significantly lowering
the glass transition temperature.
The use of fillers such as calcium carbonate, china clay
and nano-sized fillers has been extensively explored [367];
the fillers need to be well dispersed into the matrix. For
clay particles, for example, it is preferred if the particles
are fully exfoliated, although an intercalated structure is
also beneficial. With good integration between filler and
polymer, the mechanical reinforcement, thermal stabil-
ity, biodegradability and barrier properties were generally
improved. Galego et al. [368] suggested that P(3HB-co-
3HV) containing less than 8% HV content was best for a load
bearing application, because the elastic modulus reduced
dramatically above this level.
De Koning et al. [248] recognized two major prob-
lems with the higher HV content and mcl-PHAs: their low
crystallization rates and their tendency to soften and lose
their coherence at temperatures as low as 40 ◦ C makes
processing difficult. Crosslinking (such as with peroxide
addition during processing) has been tested as one solu-
tion [248]. Other additives for processing would include
slip and antiblock agents, heat stabilizers to increase the
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
processing window, antioxidants, impact modifiers, lubri-
cants to enhance flow properties for injection molding, and
melt strength enhancers for improved draw in films and
thermoforming [369].
11.6. PHA blends
The mechanical properties of blended PHA–PHA phase
separated materials have been assessed [370]. Blends of
P(3HB) with random copolymeric P(3HB-co-3HV) (contain-
ing 18.4 mol% HV and 19.7 mol% HV, with Mw of 732 and
42 kDa respectively) were prepared using a single screw
extruder followed by hot melt pressing. The 19.7 mol%
3HV copolymer blend also contained nucleating agent.
Samples crystallized from such phase separated melts
showed a higher fracture toughness than reference copoly-
mer. In fresh samples, there was a pronounced impact on
yield and fracture of the blends, with the blends being
more ductile. However, this improvement was found to be
largely lost during the ageing process. Satoh et al. [371]
studied the hydrolysis of P(3HB)/(3HB-co-3HV) (22.3 mol%
3HV) blends. DSC results showed that P(3HB) was the
main component of the crystalline phase, and the crys-
tallinity decreased with increasing wt% P(3HB-co-3HV) in
the blends. There was a correlation between an accelerated
enzymatic degradation rate and a decrease of crystallinity
in the blends.
The blending of PHA with other polymeric materials
is a technique for modifying end-use properties that is
very cost effective and relatively easy to adopt. P(3HB-
co-3HV) is miscible with many polymers such as poly-
ethylene oxide, polyvinyl alcohol, poly (p-vinylphenol),
poly (epichlorohydrin-co-ethylene oxide) and poly(vinyl
acetate) (displaying a single glass transition temperature).
P(3HB-co-3HV) is immiscible with a great many other
materials, including (as examples) polybutylene adipate,
polylactic acid > 20 kDa, ethylene-vinyl acetate copoly-
mer (containing 28 mol% vinyl acetate), poly(ethylene
succinate), poly(propylene carbonate), poly(butylene suc-
cinate), and poly(vinylidene fluoride) (see [19,372,373]
and references therein). The use of multilayer and fiber-
reinforced composites is also a valuable approach that can
lead to property improvement while lowering overall cost
[373].
11.7. PHA processing
The processing of PHA polymers is also an area that
requires careful management, particularly considering the
widely reported (although not necessarily intrinsic) [374]
thermal instability of such polymers. The use of reverse
temperature profile processing [375] is an important tech-
nique that is commonly used to minimize degradation with
melt extrusion. The unique rheological performance of PHA
materials (which have very low viscosity at melt and thus
can be used to produce articles with fine detail) is another
important aspect that cannot be covered herein [365].
Overall, it can be seen that a specific PHB or PHBV used
for one investigation can exhibit quite different properties
than that from another study, due to differences in ther-
mal stability, impurities, blend composition or copolymer
427
heterogeneities and processing conditions. Care therefore
needs to be taken that the PHA material being used in a
particular study is well-characterized before extrapolating
conclusions to other systems.
12. Outlook
Polyhydroxyalkanoates have been intensively studied
for some decades and are very attractive candidates as
supplementary or replacement materials for petroleum-
derived polyolefins, having good general properties and
being truly biodegradable. Many of the limitations that
were identified early on have been addressed through the
use of different copolymeric compositions and combina-
tions to improve mechanical properties and also through
the addition of appropriate additives and advanced
processing and modeling. Mixed culture PHA production is
emerging as an attractive approach that has the potential
to use cheap waste materials and low-capital installations
for the production of cost-effective PHA. However, in spite
of the wealth of experience embodied in the reviewed sci-
entific literature, core fundamental principles still need
to be developed for broader practical implementation of
PHA as bio-based engineering materials. Evidently a more
mechanistic understanding of structure function relation-
ships is required for PHA, particularly at higher comonomer
contents. Therefore it is suggested that efforts be placed on
three main fronts:
12.1. Modeling of PHA accumulation process and chain
growth kinetics
There is a need to develop predictive tools that can
model the PHA accumulation process, and in particular
provide a detailed polymer kinetic model for P(3HB-co-
3HV) chain growth, and thereby assess copolymer type and
distribution in the matrix. One of the limiting factors with
respect to developing such robust models of PHA molecular
dynamics in vivo is that there have been too few studies on
the mechanistic prediction of PHA accumulation and chain
growth under varying culture conditions. It is important
for the development of good process control that further
fundamental research into effect of process conditions,
feeding strategy and culture type on the kinetics of molec-
ular weight and compositional distribution be undertaken,
particularly in light of the evidence for progress around
block copolymer formation.
12.2. Compositional distribution and mechanical
properties
The effect of compositional distribution on macro-
molecular structure and the subsequent effect on mechan-
ical properties, secondary crystallization and other key
polymer properties, particularly for high HV content mate-
rials, is not well understood, and it is clear that even for pure
culture materials this is an important and largely unex-
plored parameter. Specific findings to date reveal that:
- Compositional distribution can be very broad in both pure
and mixed culture PHAs.
428 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
- Blends of PHBV copolymers of distinctly different com-
position can be poorly miscible or immiscible; phase
separated domains can result and can lead to unexpected
and/or uncontrolled changes in mechanical properties.
- For phase separated blends the domain that crystallizes
most rapidly apparently controls the final polymer prop-
erties.
Questions still arise in terms of the following effects:
- What impact does a broad compositional distribution
have on end use properties and crystallization rates?
- How predictable are the properties once compositional
distribution is characterized, and can properties be tailo-
red by appropriate blending post extraction?
- How reproducible is the raw material?
12.3. Full characterization of mixed culture PHA
To date, there has been limited research into the link
between compositional distribution, process parameters,
feeding strategies and final mechanical properties for
mixed culture PHA. It is important to demonstrate that the
chemomechanical properties are predictable and may be
controlled and within desirable limits for specific practical
applications.
In many ways these questions are all interlinked and
feed in to the full integration of process modeling with
product performance for reliable and cost effective PHA
polymer production. Clearly there is great potential and
much growing anticipation for the broad application of PHA
materials.
Acknowledgments
This research was supported under Australian Research
Council’s Linkage Projects funding scheme (project number
LP099091). The authors thank Paul Luckman for prepara-
tion of Figs. 1, 3 and 9, and Monica Arcos-Hernandez for
valuable discussions relating to PHA in general.
Appendix A. Physical properties of mixed culture PHA

Culture enrichment PHA accumulation PHA composition PHA content Thermal properties Molecular Ref.
weight data

Substrate System Substrate VFA (g%) mol% (%) (Max) Tm(1) (◦ C) Tm(2) (◦ C) Total Tg1 (◦ C) Tg2 (◦ C) PDI Mw (×105 )
HAc:HPr:HBut: 3HB:3H2MB:3HV: Hm
HVal:other 3H2MV:3HHx (J g−1 )

Municipal wastewater AE EBPR Acetate 100:0:0:0:0 75:0:25:0:0 19.2 – – – – – 1.9 6.5 [348]
Municipal wastewater AE EBPR Propionate 0:100:0:0:0 28:0:72:0:0 14.9 – – – – – 2.1 6 [348]
Municipal wastewater AE EBPR Butyrate 0:0:100:0:0 60:0:40:0:0 11.4 – – – – – 1.6 4 [348]
Municipal wastewater AE SBR Butyrate 0:0:100:0:0 100:0:0:0:0 37 178 – – – – – – [376]

B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442

Municipal wastewater AE SBR Butyrate/valerate 0:0:80:20:0 88:0:12:0:0 40 144 – – – – – – [376]
Municipal wastewater AE SBR Butyrate/valerate 0:0:60:40:0 70:0:30:0:0 35 133 – – – – – – [376]
Municipal wastewater AE SBR Butyrate/valerate 0:0:40:60:0 65:0:35:0:0 24 127 – – – – – – [376]
Municipal wastewater AE SBR Butyrate/valerate 0:0:20:80:0 49:0:51:0:0 18 109 – – – – – – [376]
Municipal wastewater AE SBR Valerate 0:0:0:100:0 46:0:54:0:0 22 99 – – – – – – [376]
Acetate/propionate AE Acetate/propionate Unspecified 90:0:10:0:0 – 168 – 44.8 1.6 – 3.3 4 [46]
Acetate/propionate AE Acetate/propionate Unspecified 75:0:25:0:0 – 139 – 4.7 56 – 2.5 17 [46]
Acetate/propionate AE Acetate/propionate Unspecified 70:0:30:0:0 – 141 – 4.7 42 – 2.1 18 [46]
Acetate AE Acetate 100:0:0:0:0 100:0:0:0:0 – 145 – 40 −19 – 1.2 3.5 [46]
Acetate/propionate MAA/AE Acetate 100:0:0:0:0 58:0:42:0:0 35.6 75 – – – – – 3.23 [377]
(P
limited)
Acetate AN/AE Acetate (Ae) 100:0:0:0:0 93:0:7:0:0 41 – – – – – 2.7 1.9 [44]
Acetate AN/AE Acetate (Ae) 100:0:0:0:0 89:0:11:0:0 – – – – – – 2.6 1.1 [44]
Acetate AN/AE Acetate (Ae) 100:0:0:0:0 80:0:16:4:0 – 170 – 25.1 −3 – 2.2 3.1 [44]
Acetate AN/AE Acetate (Ae) 100:0:0:0:0 77:0:23:0:0 – – – – – – 3.7 3.2 [44]
Acetate AN/AE Acetate (An) 100:0:0:0:0 63:0:25:12:0 – 126 – 14.5 −4.8 – 1.6 3.9 [44]
Acetate AN/AE Acetate (An) 100:0:0:0:0 71:0:29:0:0 – 112 – 6.6 −3.7 – 1.4 4.1 [44]
Acetate AN/AE Acetate (An) 100:0:0:0:0 68:0:32:0:0 – 99 – 11.9 −6.6 – 1.4 5.5 [44]
Acetate AN/AE Acetate (An) 100:0:0:0:0 65:0:35:0:0 28 105 – 44.9 −9.7 – 5.6 2.9 [44]
Acetate AN/AE Acetate (An) 100:0:0:0:0 65:0:23:12:0 – 96 – 19.8 0.3 – 1.9 1.4 [44]
Acetate AN/AE Acetate (An) 100:0:0:0:0 70:0:30:0:0 31 106 – 37.0 −3.9 – 2.9 2.5 [44]
Acetate AN/AE Propionate 0:100:0:0:0 11:13:35:41:0 – 83.8 20 – −0.2 – 1.5 5.6 [73]
Acetate AN/AE Acetate/propionate 12.5:87.5:0:0:0 20:9:39:32:0 – 119 154.9 9.5 −6.1 – 3.2 4.1 [73]
Acetate AN/AE Acetate/propionate 36.4:63.6:0:0:0 46:8:32:15:0 – 70.3 96.5 18.5 −1.6 – 1.7 5 [73]
Acetate AN/AE Acetate/propionate 69.6:30.4:0:0:0 55:5:30:9:0 – 124.3 156.4 31.5 −8.1 – 3.1 3.9 [73]
Acetate AN/AE Acetate 100:0:0:0:0 66:4:28:2:0 – 96.4/109.1 145.5/161 9.1 −6.1 – 1.4 5.5 [73]
Acetate EA SBR Acetate/propionate 55:45:0:0:0 54:0:33:13:0 26 – – – – – 2.98 29.8 [378]
Acetate EA SBR Propionate 0:100:0:0:0 31:0:47:22:0 13.5 – – – – – 2.34 23.4 [378]
Acetate EA SBR Propionate 0:100:0:0:0 12:0:67:21:0 23.7 – – – – – 2.27 22.7 [378]
Acetate EA SBR Butyrate 0:0:100:0:0 100:0:0:0:0 14.8 – – – – – 2.79 27.9 [378]
Acetate EA SBR Valerate 0:0:0:100:0 32:0:52:16:0 14.6 – – – – – 2.32 23.2 [378]
Propionate EA SBR Propionate 0:100:0:0:0 28:0:72:0:0 20.8 99.2 – 8.9 −29.6 – 2.87 28.7 [378]

429
 430
Appendix A (Continued)

Culture enrichment PHA accumulation PHA composition PHA content Thermal properties Molecular Ref.
weight data

Substrate System Substrate VFA (g%) mol% (%) (Max) Tm(1) (◦ C) Tm(2) (◦ C) Total Tg1 (◦ C) Tg2 (◦ C) PDI Mw (×105 )
HAc:HPr:HBut: 3HB:3H2MB:3HV: Hm
HVal:other 3H2MV:3HHx (J g−1 )

Acetate AE SBR Acetate 100:0:0:0:0 100:0:0:0:0 21 171 – 63.6 – – 1.3 31.6 [349]
(N
limited)
Acetate/propionate AE SBR Acetate/propionate 59:41:0:0:0 94:0:6:0:0 25 157 – 53.6 1 – 1.3 32.7 [349]
(N
limited)
Acetate AE SBR Mixed VFA – 31:18:29:22:0 61:0:39:0:0 77 113 138 – −16 – 2.3 2.15 [61]
simulated feed

B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442

Acetate AE SBR Mixed VFA – 60:16:20:4:0 79:0:21:0:0 68 121 137 – −10 – 2.7 3.86 [61]
simulated feed
Acetate AE SBR Fermented 60:9:25:6:0 85:0:15:0:0 56 134 147 – −1 – 2.3 6.46 [61]
molasses
Fermented molasses AN/AE Fermented 75:8:13:5:0 70:0:25:0:5 37 129.2 144.5 22.8 −7.2 0.5 2.1 4 [62]
molasses
Fermented molasses AN/AE Fermented 69:21:9:1:1 56:0:43:0:1 37.3 155 – 21.1 −5.5 1.2 1.8 3.5 [62]
molasses
Fermented molasses AN/AE Fermented 72:6:20:3:0 60:2:13:1:23 31.6 – – – −3.3 – 1.8 4.3 [62]
molasses
Fermented molasses AN/AE Fermented 71:7:17:3:2 47:0:53:0:0 – 153.7 165.3 10.2 −10.3 2.4 – – [62]
molasses
Fermented molasses AN/AE Acetate 100:0:0:0:0 90:4:4:1:1 29 170.9 – 77 4.6 – 2 8.1 [62]
Fermented molasses AN/AE Acetate 100:0:0:0:0 96:0:3:1:1 – 174 – 82.1 – – – – [62]
Fermented molasses AN/AE Acetate 100:0:0:0:0 89:0:11:0:0 – 164.6 171.9 77.5 4.8 – – – [62]
Fermented molasses AN/AE Propionate 0:100:0:0:0 12:6:63:14:6 15.8 89.2 – 17.6 −14.0 – 2 4.5 [62]
Fermented molasses AN/AE Butyrate 0:0:100:0:0 83:5:7:2:2 21.7 137.4 149.9 50.4 2.8 – 1.7 9 [62]
Fermented molasses AN/AE Valerate 0:0:0:100:0 12:5:78:4:1 24.8 98.1 – 1.5 −12.3 – 3.9 6.2 [62]
Fermented WAS 12 AE SBR Fermented WAS 42:18:17:3:20 82:0:18:0:0 8 131 143 42 0.7 – 2.6 6.1 [379]
gCOD/L/day
Fermented WAS 12 AE SBR Mixed VFA – 43:16:14:6:21 87:0:13:0:0 29 131 144 46 0.1 – 2.5 8.4 [379]
gCOD/L/day simulated feed
Fermented WAS 12 AE SBR Mixed VFA – 43:16:14:6:21 85:0:15:0:0 13 136 148 36 1.9 – 2.1 9.2 [379]
gCOD/L/day simulated feed
Fermented WAS 6 AE SBR Fermented WAS 42:18:17:3:20 74:0:26:0:0 19 138 147 25 −0.9 – 1.8 8.2 [379]
gCOD/L/day
Fermented WAS 6 AE SBR Mixed VFA – 43:16:14:6:21 71:0:29:0:0 23 140 155 16 −0.9 – 1.5 10.4 [379]
gCOD/L/day simulated feed
Fermented WAS 6 AE SBR Mixed VFA – 43:16:14:6:21 56:0:44:0:0 23 139 151 32 −13.2 2.2 2.5 8.8 [379]
gCOD/L/day simulated feed
Fermented WAS 6 AE SBR Acetate/propionate 77:23:0:0:0 72:0:28:0:0 23 146 – 17 −1.2 – 2.2 7.3 [379]
gCOD/L/day
Fermented WAS 6 AE SBR Acetate/propionate 77:23:0:0:0 73:0:27:0:0 25 142 – 14 −1.4 – 2.4 8.7 [379]
gCOD/L/day
Glycerol AE Glycerol Glycerol 100:0:0:0:0 10–35 158–175 – 85–95 −5 to 5 – – 2.0–3.8 [351]
AE = aerobic; AN = anaerobic; COD = chemical oxygen demand; EBPR = enhanced biological phosphorus removal; HAc = acetate; HPr = propionate; HBut = butyrate; HVal = valerate; MAA = microaerophilic;
SBR = sequencing batch reactor.
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442 431

Appendix B. Effect of polymer composition on mechanical properties of PHA – data sampled from the literature

Sample type Aging time mol% Tm (◦ C) Tg (◦ C) Young’s Stress at Tensile Elongation Ref.
comonomer modulus yield (MPa) strength at break (%)
unit (GPa) (MPa)

P(3HB)
Melt pressed sheet 1 week 0 1.5–1.8 8–20 0.8 [358]
Solvent cast film Unspecified 0 30 4 [221]
Solvent cast film 4 weeks 0 178.2 2.4 1.5 1.93 [380]
Injection molded 150 days (1% boron nitride) 0 3.5 7 [381]
Solvent cast film Unspecified 0 177 2 43 5 [24]
Solvent cast film Unspecified 0 177 4 43 5 [382]
Solvent cast film 6 days dried, hot pressed 140 ◦ C 0 172 ∼0 1.2 25 [383]
Melt pressed sheet 3 weeks 0 181 17.7 2.99 36.4 2.1 [384]
Melt pressed sheet Small amount of talc, 60 ◦ C 24 h 0 179 10 3.5 40 [22]
Unspecified Unspecified 0 175 −4 4 40 6 [385]
Unspecified Unspecified 0 175 15 40 6 [386]
Unspecified Unspecified 0 177 3.5 40 8 [264]
Solvent cast film 60 ◦ C 20 h 0 175 9 3.6 44 3 [387]
Solvent cast film 60 days 0 32.7 3.3 [388]
Solvent cast film Unspecified 0 162 −1.2 1.51 18.5 4.45 [346]
Solvent cast film “Several days” 0 2.1 28 3.6 [389]
P(3HB-co-3HV)
Film, unspecified Unspecified 0.05 172 3.0 6.0 [199]
preparation
Melt pressed sheet 24 ◦ C 1 day 1.1 158 3.65 19.7 0.17 [390]
Melt sheet Small amount of talc, 60 ◦ C 24 h 3 170 8 2.9 38 [22]
Solvent cast film 40 ◦ C, >1 day, high vacuum 3.6 169.6 3.18 50.5 1.91 [391]
Solvent cast film 25 ◦ C 2 weeks; 60 ◦ C 2 weeks 5 166 2 [392]
under vacuum
Melt pressed sheet 25 ◦ C 2 weeks; 60 ◦ C 2 weeks 5 171 −0.4 13 22.1 1.39 [392]
under vacuum
Melt pressed sheet Unspecified 5 166 1 1.4 19 2.4 [393]
Melt pressed sheet 50 ◦ C 2 days; 25 ◦ C 2 weeks 5 171 0.95 34.5 4 [394]
Solvent cast film Unspecified 5 170.1 2.3 1.47 18.09 2.31 [395]
Solvent cast film Unspecified 5 156 −1.7 2.4 123 [346]
Solvent cast film 40 ◦ C, 2 days 6.6 1.37 26.9 4.1 [396]
Solvent cast film Unspecified 7 0.124 3.0 [397]
Injection molded 90 ◦ C 12 h 8 153 3.5 27.1 3.6 [398]
dogbones
Melt pressed sheet 145 ◦ C 1 h then slow cooled 8 1.06 64 [344]
Solvent cast film 224 days 8 164 −0.6 21.5 16 [388]
Solvent cast film Unspecified 8 151 10 0.187 19 15 [297]
Melt pressed sheet Small amount of talc, 60 ◦ C 24 h 9 162 6 1.9 37 [22]
Unspecified Unspecified 10 140 1.2 25 20 [264]
Solvent cast film 224 days 10 163.2 1.7 19.3 25 [388]
Solvent cast film Unspecified 10 145 1.5 27 1 [340]
Solvent cast film 60 ◦ C 20 h 11.3 157 2 3.2 38 5 [387]
Melt pressed sheet 25 ◦ C 2 weeks; 60 ◦ C 2 weeks 12 143 −2.7 8.67 20.9 4.49 [392]
under vacuum
Melt pressed sheet Unspecified 13 157 2.9 1.07 12.7 [399]
Injection molded Unspecified 13 157.3 0.3 1 26 [400]
Solvent cast film 224 days 13 161 −0.9 18.8 48 [388]
Solvent cast film Unspecified 14 0.082 1.03 [397]
Melt pressed sheet Small amount of talc, 60 ◦ C 24 h 14 150 4 1.5 35 [22]
Solvent cast film 60 ◦ C 20 h 15.5 140 −1 2.5 34 30 12 [387]
Solvent cast film Unspecified 17 21 203 [221]
Solvent cast film 224 days 17 124.4 −4.4 18.2 43 [388]
Spin cast films 40 ◦ C, 4 days 18 158/172 13 1.12 25.1 4.03 [401]
Solvent cast film 224 days 18 123 −1.9 20.2 164 [388]
Melt pressed sheet 25 ◦ C 2 weeks; 60 ◦ C 2 weeks 20 132 −0.05 8.37 15.2 1.26 [392]
under vacuum
Solvent cast film 3 weeks 20 145 −1 2.9 50 [24]
Melt pressed sheet Small amount of talc, 60 ◦ C 24 h 20 145 −1 1.2 32 [22]
Unspecified Unspecified 20 130 0.8 20 50 [264]
Solvent cast film 224 days 20 116.4 −6.3 15.6 182 [388]
Solvent cast film 60 ◦ C 20 h 20.1 114 −5 1.9 26 23 27 [387]
Solvent cast film Unspecified 22 0.09 2.29 [397]
Film, unspecified Unspecified 22.6 97 −6.4 4.0 [199]
preparation
◦
Melt pressed sheet Small amount of talc, 60 C 24 h 25 137 −6 0.7 30 [22]
Solvent cast film 60 ◦ C 20 h 28.4 102 −8 1.4 22 30 700 [387]
432 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442

Sample type Aging time mol% Tm (◦ C) Tg (◦ C) Young’s Stress at Tensile Elongation Ref.
comonomer modulus yield (MPa) strength at break (%)
unit (GPa) (MPa)

Solvent cast film Unspecified 29 87 −10 0.144 15 900 [297]

Solvent cast film Unspecified 34 97 −8 13 18 970 [24]
Melt pressed sheet Unspecified 36 166 ∼0 0.5 16.6 70 [383]
Film, unspecified Unspecified 39.2 70 −5.8 3.2 [199]
preparation
Melt pressed sheet Unspecified 50 162 ∼0 0.41 13.4 230 [383]
Film, unspecified Unspecified 54.1 64 −6.4 2.6 [199]
preparation
Solvent cast film Unspecified 55 77 −10 12 16 >1200 [24]
Film, unspecified Unspecified 69.1 70 −3.9 1.8 [199]
preparation
Solvent cast film Unspecified 71 83 −13 11 5 [24]
Solvent cast film Unspecified 25–30 18 120 [221]
P(3HB-co-4HB)
Films (solvent cast) 3 weeks 3 166 28 45 [325]
Films (solvent cast) 4 weeks 5 168.5 −2 1.23 10.7 [380]
Films (solvent cast) 3 weeks 10 159 24 242 [325]
Films (solvent cast) 3 weeks 16 150 −7 26 444 [325]
Films (solvent cast) 4 weeks 24 161.1 −5 0.79 22.2 [380]
Films (solvent cast) 4 weeks 38 152.2 −10 0.66 48 [380]
Films (solvent cast) 3 weeks 44 10 511 [325]
Films (solvent cast) 3 weeks 64 50 −35 0.03 17 591 [402]
Films (solvent cast) 3 weeks 78 49 −37 0.024 42 1120 [402]
Films (solvent cast) 3 weeks 82 52 −39 0.045 58 1320 [402]
Films (solvent cast) 3 weeks 90 50 −42 0.1 65 1080 [402]
Films (solvent cast) Unspecified 94 51 −46 0.055 39 500 [403]
Films (solvent cast) 3 weeks 100 53 −48 0.149 104 1000 [402]
P(3HB-co-3H4MV)
Films (solvent cast) 180 days 7 149, 162 −2 18 19 [404]
Films (solvent cast) 180 days 9 146, 159 −2 25 22 [404]
Films (solvent cast) 180 days 14 136, 148 −2 19 110 [404]
Films (solvent cast) 180 days 16 133, 143 −2 17 260 [404]
Films (solvent cast) 180 days 22 126 −2 14 440 [404]
P(3HB-co-3HHx)
Melt pressed sheet Unspecified 2.5 142 0.631 25.7 6.7 [324]
Melt pressed sheet 3 months 2.5 142 5 [324]
Melt pressed sheet 3 months 4.5 22 [324]
Melt pressed sheet Unspecified 4.6 130 0.6 22.9 6.5 [324]
Melt pressed sheet 3 months 4.7 3 [324]
Melt pressed sheet Unspecified 5.4 0.494 23.9 17.6 [324]
Melt pressed sheet 3 months 5.4 130 10 [324]
Melt pressed sheet Unspecified 5.9 130 0.412 17.7 163 [324]
Melt pressed sheet Unspecified 6.3 0.343 17.9 24 [324]
Melt pressed sheet 3 months 6.4 128 6 [324]
Melt pressed sheet Unspecified 7 0.289 17.3 23.6 [324]
Melt pressed sheet 3 months 7.6 119 12 [324]
Melt pressed sheet 3 months 8.1 119 12 [324]
Melt pressed sheet Unspecified 8.4 0.989 14.5 11.9 [324]
Melt pressed sheet Unspecified 8.5 0.249 17.8 25.1 [324]
Melt pressed sheet Unspecified 8.5 0.232 15.6 34.3 [324]
Melt pressed sheet Unspecified 9.5 0.155 8.8 43 [324]
Melt pressed sheet Unspecified 10 ∼0.52 [38]
Films (solvent cast) Unspecified 10 127 −1 21 400 [382]
Films (solvent cast) Unspecified 12 97 −1.8 0.135 4.45 107.7 [346]
Films (solvent cast) Unspecified 12 96.2 −1.2 0.498 9.36 112.9 [395]
Films (solvent cast) Unspecified 15 115 0 23 760 [382]
Films (solvent cast) Unspecified 17 120 −2 20 850 [382]
Melt pressed sheet Unspecified ∼12 106.3 −0.2 0.267 5.99 11.5 [405]
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442 433

References [26] Anderson AJ, Haywood GW, Dawes EA. Biosynthesis and composi-
tion of bacterial poly(hydroxyalkanoates). International Journal of
Biological Macromolecules 1990;12:102–5.
[1] Rehm BHA. Polyester synthases: natural catalysts for plastics. Bio-
[27] Lemoigne M. Production d’acide ␤-oxybutyrique par certaines bac-
chemical Journal 2003;376:15–33.
téries du groupe du B. subtilis. Comptes Rendus de l’Académie des
[2] Anderson AJ, Dawes EA. Occurrence, metabolism, metabolic role,
Sciences 1923;176:1761–3.
and industrial uses of bacterial polyhydroxyalkanoates. Microbio-
[28] Lemoigne M. Sur le mécanisme de la production de l’acide ␤-
logical Reviews 1990;54:450–72.
oxybutyrique par voie biochimique. Comptes Rendus de l’Académie
[3] Pötter M, Steinbüchel A. Poly(3-hydroxybutyrate) granule-
des Sciences 1924;178:1093–5.
associated proteins: impacts on poly(3-hydroxybutyrate)
[29] Lemoigne M. Sur l’origine de l’acide ␤-oxybutyrique obtenu par
synthesis and degradation. Biomacromolecules 2005;6:
processus microbien. Comptes Rendus de l’Académie des Sciences
552–60.
1925;180:1539–41.
[4] Steinbüchel A, Valentin HE. Diversity of bacterial polyhydroxyalka-
[30] Lemoigne M. Études sur I’autolyse microbienne. Acidification par
noic acids. FEMS Microbiology Letters 1995;128:219–28.
formation d’acide ␤-oxybutyrique. Annales de l’Institut Pasteur
[5] Steinbüchel A, Lutke-Eversloh T. Metabolic engineering and path-
Paris 1925;39:l44–173.
way construction for biotechnological production of relevant
[31] Lemoigne M. Origine de l’acide 3-oxo-butyrique obtenue par autol-
polyhydroxyalkanoates in microorganisms. Biochemical Engineer-
yse microbienne. Comptes Rendus des Seances de la Societe de
ing Journal 2003;16:81–96.
Biologie et de Ses Filiales 1926;95:1359–60.
[6] Kim YB, Lenz RW. Polyesters from microorganisms. Advances in
[32] Lemoigne M. Produit de déshydratation et de polymérisation de
Biochemical Engineering/Biotechnology 2001;71:51–79.
l’acide ␤-oxybutyrique. Bulletin de la Societe de Chimie Biologique
[7] Lenz RW, Marchessault RH. Bacterial polyesters: biosynthesis,
1926;8:770–82.
biodegradable plastics and biotechnology. Biomacromolecules
[33] Lemoigne M. Études sur l’autolyse microbienne. Origine de l’acide
2005;6:1–8.
␤-oxybutyrique formé par autolyse. Annales de l’Institut Pasteur
[8] Hazer B, Steinbüchel A. Increased diversification of polyhy-
Paris 1927;41:148–65.
droxyalkanoates by modification reactions for industrial and
[34] Williamson DH, Wilkinson JF. The isolation and estimation of the
medical applications. Applied Microbiology and Biotechnology
poly-␤-hydroxybutyrate inclusions of bacillus species. Journal of
2007;74:1–12.
General Microbiology 1958;19:198–209.
[9] Sudesh K, Abe H. Practical guide to microbial polyhydroxyalka-
[35] Doudoroff M, Stanier RY. Role of poly-␤-hydroxybutyric acid in
noates. Shrewsbury, UK: Smithers Rapra Technology; 2010, 160
the assimilation of organic carbon by bacteria. Nature 1959;183:
pp.
1440–2.
[10] Rehm BHA. Bacterial polymers: biosynthesis, modifications and
[36] Baptist JN, Ziegler JB. Method of making absorbable surgical sutures
applications. Nature Reviews Microbiology 2010;8:578–92.
from poly ␤ hydroxy acids. US Patent 3,225,766 (A), W R Grace &
[11] Chanprateep S. Current trends in biodegradable polyhy-
Co.; 1965.
droxyalkanoates. Journal of Bioscience and Bioengineering
[37] Holmes PA. Applications of PHB – a microbially produced
2010;110:621–32.
biodegradable thermoplastic. Physics in Technology 1985;16:32–6.
[12] Chen G. Plastics completely synthesized by bacteria: polyhydrox-
[38] Noda I, Green PR, Satkowski MM, Schechtman LA. Preparation and
yalkanoates. In: Chen G, editor. Plastics from bacteria: natural
properties of a novel class of polyhydroxyalkanoate copolymers.
functions and applications. Berlin; Heidelberg: Springer-Verlag;
Biomacromolecules 2005;6:580–6.
2010. p. 17–37.
[39] Byrom D. Polyhydroxyalkanoates. In: Mobley DP, editor. Plastics
[13] Verlinden RAJ, Hill DJ, Kenward MA, Williams CD, Radecka I. Bacte-
from microbes: microbial synthesis of polymers and polymer pre-
rial synthesis of biodegradable polyhydroxyalkanoates. Journal of
cursors. Munich: Hanser; 1994. p. 5–33.
Applied Microbiology 2007;102:1437–49.
[40] Dias JML, Lemos PC, Serafim LS, Oliveira C, Eiroa M, Albuquerque
[14] Pan P, Inoue Y. Polymorphism and isomorphism in biodegradable
MGE, Ramos AM, Oliveira R, Reis MA. Recent advances in poly-
polyesters. Progress in Polymer Science 2009;34:605–40.
hydroxyalkanoate production by mixed aerobic cultures: from
[15] Marchessault RH. Polyhydroxyalkanoate (PHA) history at Syracuse
the substrate to the final product. Macromolecular Bioscience
University and beyond. Cellulose 2009;16:357–9.
2006;6:885–906.
[16] Orts WJ, Nobes GAR, Kawada J, Nguyen S, Yu G-E, Raveneile F.
[41] Serafim LS, Lemos PC, Albuquerque MGE, Reis MAM. Strategies for
Poly(hydroxyalkanoates): biorefinery polymers with a whole range
PHA production by mixed cultures and renewable waste materials.
of applications. The work of Robert H. Marchessault. Canadian Jour-
Applied Microbiology and Biotechnology 2008;81:615–28.
nal of Chemistry 2008;86:628–40.
[42] Albuquerque MGE, Eiroa M, Torres C, Nunes BR, Reis MAM.
[17] Castilho LR, Mitchel DA, Freire DMG. Production of polyhydrox-
Strategies for the development of a side stream process for poly-
yalkanoates (PHAs) from waste materials and by-products by
hydroxyalkanoate (PHA) production from sugar cane molasses.
submerged and solid-state fermentation. Bioresource Technology
Journal of Biotechnology 2007;130:411–21.
2009;100:5996–6009.
[43] Bengtsson S, Werker A, Christensson M, Welander T. Production of
[18] Khanna S, Srivastava AK. Recent advances in microbial polyhydrox-
polyhydroxyalkanoates by activated sludge treating a paper mill
yalkanoates. Process Biochemistry 2005;40:607–19.
wastewater. Bioresource Technology 2008;99:509–16.
[19] Sudesh K, Abe H, Doi Y. Synthesis, structure and properties of
[44] Dai Y, Yuan ZG, Jack K, Keller J. Production of targeted
polyhydroxyalkanoates: biological polyesters. Progress in Polymer
poly(3-hydroxyalkanoates) copolymers by glycogen accumulating
Science 2000;25:1503–55.
organisms using acetate as sole carbon source. Journal of Biotech-
[20] Braunegg G, Lefebvre G, Genser KF. Polyhydroxyalkanoates,
nology 2007;129:489–97.
biopolyesters from renewable resources: physiological and engi-
[45] Kleerebezem R, van Loosdrecht MCM. Mixed culture biotechnol-
neering aspects. Journal of Biotechnology 1998;65:127–61.
ogy for bioenergy production. Current Opinion in Biotechnology
[21] Amass W, Amass A, Tighe B. A review of biodegradable polymers:
2007;18:207–12.
uses, current developments in the synthesis and characterization
[46] Reis MAM, Serafim LS, Lemos PC, Ramos AM, Aguiar FR,
of biodegradable polyesters, blends of biodegradable polymers and
van Loosdrecht MCM. Production of polyhydroxyalkanoates by
recent advances in biodegradation studies. Polymer International
mixed microbial cultures. Bioprocess and Biosystems Engineering
1998;47:89–144.
2003;25:377–85.
[22] Holmes PA. Biologically produced (R)-3-hydroxyalkanoate poly-
[47] Gurieff N, Lant P. Comparative life cycle assessment and finan-
mers and copolymers. In: Bassett OC, editor. Developments in
cial analysis of mixed culture polyhydroxyalkanoate production.
crystalline polymers – 2. Essex, UK: Elsevier Applied Science Pub-
Bioresource Technology 2007;98:3393–403.
lishers Ltd.; 1988. p. 1–65.
[48] Wallen LL, Rohwedder WK. Poly-␤-hydroxyalkanoate from
[23] Kessler B, Witholt B. Synthesis, recovery and possible application of
activated sludge. Environmental Science and Technology
medium-chain-length polyhydroxyalkanoates: a short overview.
1974;8:576–9.
Macromolecular Symposium 1998;130:245–60.
[49] Daigger GT, Grady CPL. The dynamics of microbial growth
[24] Doi Y. Microbial polyesters. New York: VCH Publishers; 1990, 166
on soluble substrates – a unifying theory. Water Research
pp.
1982;16:365–82.
[25] Sudesh K, Doi A. Polyhydroxyalkanoates. In: Bastioli C, editor.
[50] Salehizadeh H, Van Loosdrecht MCM. Production of polyhydrox-
Handbook of biodegradable polymers. United Kingdom: RAPRA
yalkanoates by mixed culture: recent trends and biotechnological
Technology Ltd.; 2005. p. 219–56.
importance. Biotechnology Advances 2004;22:261–79.
434 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
[51] Beccari M, Majone M, Massanisso P, Ramadori R. A bulking sludge
with high storage response selected under intermittent feeding.
Water Research 1998;32:3403–13.
[52] Beun JJ, Dircks K, Heijnen JJ, van Loosdrecht MCM. Poly-␤-
hydroxybutyrate metabolism in dynamically fed mixed microbial
cultures. Water Research 2002;36:1167–80.
[53] Bengtsson S, Pisco AR, Reis MAM, Lemos PC. Production of polyhy-
droxyalkanoates from fermented sugar cane molasses by a mixed
culture enriched in glycogen accumulating organisms. Journal of
Biotechnology 2010;145:253–63.
[54] Pisco AR, Bengtsson S, Werker A, Reis MAM, Lemos PC.
Community structure evolution and enrichment of glycogen-
accumulating organisms producing polyhydroxyalkanoates from
fermented molasses. Applied and Environmental Microbiology
2009;75:4676–86.
[55] Dionisi D, Majone M, Papa V, Beccari M. Biodegradable poly-
mers from organic acids by using activated sludge enriched
by aerobic periodic feeding. Biotechnology and Bioengineering
2004;85:569–79.
[56] Beccari M, Dionisi D, Giuliani A, Majone M, Ramadori R. Effect of
different carbon sources on aerobic storage by activated sludge.
Water Science and Technology 2002;45:157–68.
[57] Serafim LS, Lemos PC, Oliveira R, Reis MAM. Optimization of polyhy-
droxybutyrate production by mixed cultures submitted to aerobic
dynamic feeding conditions. Biotechnology and Bioengineering
2004;87:145–60.
[58] Lemos PC, Serafim LS, Reis MAM. Synthesis of polyhydroxyalka-
noates from different short-chain fatty acids by mixed cultures
submitted to aerobic dynamic feeding. Journal of Biotechnology
2006;122:226–38.
[59] Solaiman DKY, Ashby RD, Foglia TA, Marmer WM. Con-
version of agricultural feedstock and coproducts into
poly(hydroxyalkanoates). Applied Microbiology and Biotechnology
2006;71:783–9.
[60] Albuquerque MGE, Bengtsson S, Martino V, Pollet E, Reis MAM. Eco-
engineering of mixed microbial cultures to develop a cost-effective
bioplastic (PHA) production process from a surplus feedstock –
sugar molasses. Journal of Biotechnology 2010;150:S70.
[61] Albuquerque MGE, Martino V, Pollet E, Avérous L, Reis MAM. Mixed
culture polyhydroxyalkanoate (PHA) production from volatile fatty
acid (VFA)-rich streams: effect of substrate composition and feed-
ing regime on PHA productivity, composition and properties.
Journal of Biotechnology 2011;151:66–76.
[62] Bengtsson S, Pisco AR, Johansson P, Lemos PC, Reis MAM. Molecular
weight and thermal properties of polyhydroxyalkanoates produced
from fermented sugar molasses by open mixed cultures. Journal of
Biotechnology 2010;147:172–9.
[63] Md Din MF, Ujang Z, Van Loosdrecht M, Ahmad M. Polyhy-
droxyalkanoates (PHAs) production from saponified sunflower
oil in mixed cultures under aerobic condition. Jurnal Teknologi
2008;48:1–19.
[64] Gurieff N. Production of biodegradable polyhydroxyalkanoate poly-
mers using advanced biological wastewater treatment process
technology. PhD Dissertation. School of Engineering, Brisbane: The
University of Queensland; 2007. 160 pp.
[65] Liu HY, Hall PV, Darby JL, Coats ER, Green PG, Thompson DE,
Loge FJ. Production of polyhydroxyalkanoate during treatment
of tomato cannery wastewater. Water Environment Research
2008;80:367–72.
[66] Cai MM, Chua H, Zhao QL, Shirley SN, Ren J. Optimal production
of polyhydroxyalkanoates (PHA) in activated sludge fed by volatile
fatty acids (VFAs) generated from alkaline excess sludge fermenta-
tion. Bioresource Technology 2009;100:1399–405.
[67] Coats ER, Loge FJ, Wolcott MP, Englund K, McDonald AG. Synthe-
sis of polyhydroxyalkanoates in municipal wastewater treatment.
Water Environment Research 2007;79:2396–403.
[68] Dionisi D, Majone M, Levantesi C, Di Gregorio S, Beccari M. Effect
of feed length on settleability, substrate uptake and storage in a
sequencing batch reactor treating an industrial wastewater. Envi-
ronmental Technology 2006;27:901–8.
[69] Carucci A, Dionisi D, Majone M, Rolle E, Smurra P. Aerobic stor-
age by activated sludge on real wastewater. Water Research
2001;35:3833–44.
[70] Moita R, Lemos PC. Biopolymers production from mixed cul-
tures and pyrolysis by-products. Journal of Biotechnology
2012;157:578–83.
[71] Dionisi D, Carucci G, Papini MP, Riccardi C, Majone M, Carrasco F.
Olive oil mill effluents as a feedstock for production of biodegrad-
able polymers. Water Research 2005;39:2076–84.
[72] Beccari M, Bertin L, Dionisi D, Fava F, Lampis S, Majone M, Valentino
F, Vallini G, Villano M. Exploiting olive oil mill effluents as a renew-
able resource for production of biodegradable polymers through a
combined anaerobic–aerobic process. Journal of Chemical Technol-
ogy and Biotechnology 2009;84:901–8.
[73] Dai Y, Lambert L, Yuan Z, Keller J. Characterisation of poly-
hydroxyalkanoate copolymers with controllable four-monomer
composition. Journal of Biotechnology 2008;134:137–45.
[74] Dai Y, Lambert L, Yuan ZG, Keller J. Microstructure of copolymers of
polyhydroxyalkanoates produced by glycogen accumulating orga-
nisms with acetate as the sole carbon source. Process Biochemistry
2008;43:968–77.
[75] Takabatake H, Satoh H, Mino T, Matsuo T. Recovery of biodegradable
plastics from activated sludge process. Water Science and Technol-
ogy 2000;42:351–6.
[76] Lee WH, Loo CY, Nomura CT, Sudesh K. Biosynthesis of poly-
hydroxyalkanoate copolymers from mixtures of plant oils
and 3-hydroxyvalerate precursors. Bioresource Technology
2008;99:6844–51.
[77] Johnson K, Jiang Y, Kleerebezem R, Muyzer G, van Loosdrecht MCM.
Enrichment of a mixed bacterial culture with a high polyhydrox-
yalkanoate storage capacity. Biomacromolecules 2009;10:670–6.
[78] Johnson K, van Loosdrecht MCM, Kleerebezem R. Influence
of ammonium on the accumulation of polyhydroxybutyrate
(P(3HB)) in aerobic open mixed cultures. Journal of Biotechnology
2010;147:73–9.
[79] Albuquerque MGE, Martino V, Pollet E, Averous L, Reis MAM. Mixed
culture polyhydroxyalkanoate (PHA) production from volatile fatty
acid (VFA)-rich streams: effect of substrate composition and feed-
ing regime on PHA productivity, composition and properties.
Journal of Biotechnology 2011;151:66–76.
[80] Villano M, Beccari M, Dionisi D, Lampis S, Miccheli A, Vallini G.
Effect of pH on the production of bacterial polyhydroxyalkanoates
by mixed cultures enriched under periodic feeding. Process Bio-
chemistry 2010;45:714–23.
[81] Ivanova G, Serafim LS, Lemos PC, Ramos AM, Reis MAM, Cabrita
EJ. Influence of feeding strategies of mixed microbial cultures
on the chemical composition and microstructure of copolyesters
P(3HB-co-3HV) analyzed by NMR and statistical analysis. Magnetic
Resonance in Chemistry 2009;47:497–504.
[82] van Beilen JB, Poirier Y. Production of renewable polymers from
crop plants. Plant Journal 2008;54:684–701.
[83] Snell KD, Peoples OP. PHA bioplastic: a value-added coproduct
for biomass biorefineries. Biofuels, Bioproducts and Biorefining
2009;3:456–67.
[84] Poirier Y, Dennis DE, Klomparens K, Somerville C. Polyhydroxy-
butyrate, a biodegradable thermoplastic, produced in transgenic
plants. Science 1992;256:520–3.
[85] Bohmert-Tatarev K, McAvoy S, Daughtry S, Peoples OP, Snell
KD. High levels of bioplastic are produced in fertile transplas-
tomic tobacco plants engineered with a synthetic operon
for the production of polyhydroxybutyrate. Plant Physiology
2011;155:1690–708.
[86] Somleva MN, Snell KD, Beaulieu JJ, Peoples OP, Garrison BR, Pat-
terson NA. Production of polyhydroxybutyrate in switchgrass, a
value-added co-product in an important lignocellulosic biomass
crop. Plant Biotechnology Journal 2008;6:663–78.
[87] Hempel F, Bozarth AF, Lindenkamp N, Klingl A, Zauner S, Linne U,
Steinbüchel A, Maier UG. Microalgae as bioreactors for bioplastic
production. Microbial Cell Factories 2011;10:81–6.
[88] Abe H, Doi Y, Satkowski MM, Noda I. Miscibility and morphology
of blends of isotactic and atactic poly(3-hydroxybutyrate). Macro-
molecules 1994;27:50–4.
[89] Abe H, Matsubara I, Doi Y. Physical properties
and enzymatic degradability of polymer blends of
bacterial poly[(R)-3-hydroxybutyrate] and poly[(R,S)-
3-hydroxybutyrate]stereoisomers. Macromolecules
1995;28:844–53.
[90] Madison LL, Huisman GW. Metabolic engineering of poly(3-
hydroxyalkanoates): from DNA to plastic. Microbiology and
Molecular Biology Reviews 1999;63:21–53.
[91] Peoples OP, Sinskey AJ. Poly-␤-hydroxybutyrate biosynthesis in
Alcaligenes-Eutrophus H16 – characterization of the genes encoding
␤-ketothiolase and acetoacetyl-CoA reductase. Journal of Biological
Chemistry 1989;264:15293–7.
[92] Lawrence AG, Choi JH, Rha CK, Stubbe J, Sinskey AJ. In vitro analysis
of the chain termination reaction in the synthesis of poly-(R)-␤-
hydroxybutyrate by the class III synthase from Allochromatium
vinosum. Biomacromolecules 2005;6:2113–9.
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
[93] Rehm BHA, Steinbüchel A. Biochemical and genetic analysis of PHA
synthases and other proteins required for PHA synthesis. Interna-
tional Journal of Biological Macromolecules 1999;25:3–19.
[94] Grage K, Jahns AC, Parlane N, Palanisamy R, Rasiah IA, Atwood
JA, Rehm BHA. Bacterial polyhydroxyalkanoate granules: bio-
genesis, structure, and potential use as nano-/micro-beads in
biotechnological and biomedical applications. Biomacromolecules
2009;10:660–9.
[95] Ratledge C, Kristiansen B. Basic biotechnology. Cambridge, UK:
Cambridge University Press; 2001, 584 pp.
[96] Henderson RA, Jones CW. Poly-3-hydroxybutyrate production by
washed cells of Alcaligenes eutrophus; purification, characterisa-
tion and potential regulatory role of citrate synthase. Archives of
Microbiology 1997;168:486–92.
[97] Chowdhury R. Poly-␤-hydroxybuttersaüre abbauende bakterien
und exoenzyme. Archiv fur Mikrobiologie 1963;47:167–200.
[98] Nuti MP, Bertoldi Md, Lepidi AA. Influence of phenylacetic acid on
poly-␤-hydroxybutyrate (PHB) polymerization and cell elongation
in Azotobacter chroococcum Beij. Canadian Journal of Microbiology
1972;18:1257–61.
[99] Pötter M, Madkour MH, Mayer F, Steinbüchel A. Regulation
of phasin expression and polyhydroxyalkanoate (PHA) gran-
ule formation in Ralstonia eutropha H16. Microbiology-Sgm
2002;148:2413–26.
[100] Ballard DGH, Holmes PA, Senior PJ. Formation of polymers of
␤-hydroxybutyric acid in bacterial cells and a comparison of
the morphology of growth with the formation of polyethylene
in the solid state. In: Fontanille M, Guyot A, editors. Recent
advances in mechanistic and synthetic aspects of polymeriza-
tion, vol. 215. Bandol, France: NATO Asi Science Series C; 1987.
p. 293–314.
[101] Steinbüchel A, Aerts K, Babel W, Follner C, Liebergesell M, Madk-
our MH, Wieczorek R, Mayer F, Pieper-Fürst U, Pries A, Valentin HE.
Considerations on the structure and biochemistry of bacterial poly-
hydroxyalkanoic acid inclusions. Canadian Journal of Microbiology
1995;41(13):94–105.
[102] Jendrossek D, Selchow O, Hoppert M. PHB granules at the
early stages of formation are localized close to the cytoplas-
mic membrane in Caryophanon latum. Applied and Environmental
Microbiology 2007;73:586–93.
[103] Mayer F, Hoppert M. Determination of the thickness of the
boundary layer surrounding bacterial PHA inclusion bodies, and
implications for models describing the molecular architecture of
this layer. Journal of Basic Microbiology 1997;37:45–52.
[104] Dennis D, Sein V, Martinez E, Augustine B. PhaP is involved in
the formation of a network on the surface of polyhydroxyalka-
noate inclusions in Cupriavidus necator H16. Journal of Bacteriology
2008;190:555–63.
[105] Beeby M, Cho M, Stubbe J, Jensen GJ. Growth and localization of
polyhydroxybutyrate granules in Ralstonia eutropha. Journal of Bac-
teriology 2012;194:1092–9.
[106] York GM, Stubbe J, Sinskey AJ. The Ralstonia eutropha PhaR protein
couples synthesis of the PhaP phasin to the presence of polyhydrox-
ybutyrate in cells and promotes polyhydroxybutyrate production.
Journal of Bacteriology 2002;184:59–66.
[107] Neumann L, Spinozzi F, Sinibaldi R, Rustichelli F, Pötter M, Stein-
büchel A. Binding of the major phasin, PhaP1, from Ralstonia
eutropha H16 to poly(3-hydroxybutyrate) granules. Journal of Bac-
teriology 2008;190:2911–9.
[108] York GM, Stubbe J, Sinskey AJ. New insight into the role of the PhaP
phasin of Ralstonia eutropha in promoting synthesis of polyhydrox-
ybutyrate. Journal of Bacteriology 2001;183:2394–7.
[109] Tian JM, He AM, Lawrence AG, Liu P, Watson N, Sinskey AJ,
Stubbe J. Analysis of transient polyhydroxybutyrate production in
Wautersia eutropha H16 by quantitative western analysis and trans-
mission electron microscopy. Journal of Bacteriology 2005;187:
3825–32.
[110] Hanley SZ, Pappin DJC, Rahman D, White AJ, Elborough KM, Slabas
AR. Re-evaluation of the primary structure of Ralstonia eutropha
phasin and implications for polyhydroxyalkanoic acid granule
binding. FEBS Letters 1999;447:99–105.
[111] Wieczorek R, Pries A, Steinbüchel A, Mayer F. Analysis of a
24-Kilodalton protein associated with the polyhydroxyalkanoic
acid granules in Alcaligenes-Eutrophus. Journal of Bacteriology
1995;177:2425–35.
[112] Horowitz DM, Sanders JKM. Amorphous, biomimetic granules
of polyhydroxybutyrate – preparation, characterization, and bio-
logical implications. Journal of the American Chemical Society
1994;116:2695–702.
435
[113] McCool GJ, Fernandez T, Li N, Cannon MC. Polyhydroxyalkanoate
inclusion-body growth and proliferation in Bacillus megaterium.
FEMS Microbiology Letters 1996;138:41–8.
[114] Jendrossek D. Polyhydroxyalkanoate granules are complex sub-
cellular organelles (carbonosomes). Journal of Bacteriology
2009;191:3195–202.
[115] Stubbe J, Tian J. Polyhydroxyalkanoate (PHA) homeostasis: the role
of the PHA synthase. Natural Products Reports 2003;20:445–57.
[116] Cho M, Brigham CJ, Sinskey AJ, Stubbe J. Purification of polyhydrox-
ybutyrate synthase from its native organism, Ralstonia eutropha:
implications for the initiation and elongation of polymer formation
in vivo. Biochemistry 2012;51:2276–88.
[117] Rehm BHA, Qi QS, Beermann BB, Hinz HJ, Steinbüchel A. Matrix-
assisted in vitro refolding of Pseudomonas aeruginosa class II
polyhydroxyalkanoate synthase from inclusion bodies produced in
recombinant Escherichia coli. Biochemical Journal 2001;358:263–8.
[118] Jendrossek D. Fluorescence microscopical investigation of poly(3-
hydroxybutyrate) granule formation in bacteria. Biomacro-
molecules 2005;6:598–603.
[119] Gerngross TU, Reilly P, Stubbe J, Peoples OP. Immunocytochemical
analysis of poly-␤-hydroxybutyrate (PHB) synthase in Alcaligenes-
Eutrophus H16 – localization of the synthase enzyme at the surface
of PHB granules. Journal of Bacteriology 1993;175:5289–93.
[120] Tian JM, Sinskey AJ, Stubbe J. Kinetic studies of polyhy-
droxybutyrate granule formation in Wautersia eutropha H16
by transmission electron microscopy. Journal of Bacteriology
2005;187:3814–24.
[121] Mumtaz T, Abd-Aziz S, Rahman NA, Yee PL, Shirai Y, Hassan MA.
Visualization of core–shell P(3HB)V granules of wild type Coma-
monas sp. EB172 in vivo under transmission electron microscope.
International Journal of Polymer Analysis and Characterization
2011;16:228–38.
[122] Muh U, Sinskey AJ, Kirby DP, Lane WS, Stubbe JA. PHA synthase from
Chromatium vinosum: cysteine 149 is involved in covalent catalysis.
Biochemistry 1999;38:826–37.
[123] Zhang SM, Yasuo T, Lenz RW, Goodwin S. Kinetic and mechanistic
characterization of the polyhydroxybutyrate synthase from Ralsto-
nia eutropha. Biomacromolecules 2000;1:244–51.
[124] Kawaguchi Y, Doi Y. Kinetics and mechanism of synthesis and
degradation of poly(3-hydroxybutyrate) in Alcaligenes-Eutrophus.
Macromolecules 1992;25:2324–9.
[125] Jia Y, Yuan W, Wodzinska J, Park C, Sinskey AJ, Stubbe J. Mecha-
nistic studies on class I polyhydroxybutyrate (PHB) synthase from
Ralstonia eutropha: class I and III synthases share a similar catalytic
mechanism. Biochemistry 2001;40:1011–9.
[126] Wodzinska J, Snell KD, Rhomberg A, Sinskey AJ, Biemann K,
Stubbe J. Polyhydroxybutyrate synthase: evidence for covalent
catalysis. Journal of the American Chemical Society 1996;118:
6319–20.
[127] Gerngross TU, Snell KD, Peoples OP, Sinskey AJ. Overexpression and
purification of the soluble polyhydroxyalkanoate synthase from
Alcaligenes-eutrophus – evidence for a required posttranslational
modification for catalytic activity. Biochemistry 1994;33:9311–20.
[128] Tian JM, Sinskey AJ, Stubbe J. Class III polyhydroxybutyrate
synthase: involvement in chain termination and reinitiation. Bio-
chemistry 2005;44:8369–77.
[129] Cai L, Tan D, Aibaidula G, Chen J-C, Tian W-D, Chen G-Q. Compar-
ative genomics study of polyhydroxyalkanoates (PHA) and ectoine
relevant genes from Halomonas sp TD01 revealed extensive hor-
izontal gene transfer events and co-evolutionary relationships.
Microbial Cell Factories 2011;10:88.
[130] Li P, Chakraborty S, Stubbe J. Detection of covalent and non-
covalent intermediates in the polymerization reaction catalyzed
by a C149S class III polyhydroxybutyrate synthase. Biochemistry
2009;48:9202–11.
[131] Yamanaka K, Aoki T, Kudo T, Kimura Y. End-group analysis of bac-
terially produced poly(3-hydroxybutyrate): discovery of succinate
as the polymerization starter. Macromolecules 2009;42:4038–46.
[132] Doi Y. Microbial synthesis, physical properties, and biodegrad-
ability of polyhydroxyalkanoates. Macromolecular Symposium
1995;98:585–99.
[133] Kawaguchi Y, Doi Y. Kinetics and mechanism of synthesis and
degradation of poly(3-hydroxybutyrate) in Alcaligenes eutrophus.
Macromolecules 1992;25:2324–9.
[134] Kusaka S, Abe H, Lee SY, Doi Y. Molecular mass of poly[(R)-3-
hydroxybutyric acid] produced in a recombinant Escherichia coli.
Applied Microbiology and Biotechnology 1997;47:140–3.
[135] Madden LA, Anderson AJ, Shah DT, Asrar J. Chain termination
in polyhydroxyalkanoate synthesis: involvement of exogenous
436 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
hydroxy-compounds as chain transfer agents. International Journal
of Biological Macromolecules 1999;25:43–53.
[136] Kichise T, Fukui T, Yoshida Y, Doi Y. Biosynthesis of polyhydrox-
yalkanoates (PHA) by recombinant Ralstonia eutropha and effects
of PHA synthase activity on in vivo PHA biosynthesis. International
Journal of Biological Macromolecules 1999;25:69–77.
[137] Shah DT, Berger PA, Tran M, Asrar J, Madden LA, Anderson
AJ. Synthesis of poly(3-hydroxybutyrate) by Ralstonia eutropha
in the presence of 13C-labeled ethylene glycol. Macromolecules
2000;33:6624–6.
[138] Shah DT, Tran M, Berger PA, Aggarwal P, Asrar J. Synthesis
and properties of hydroxy-terminated poly(hydroxyalkanoate)s.
Macromolecules 2000;33:2875–80.
[139] Ashby RD, Solaiman DKY, Foglia TA. Synthesis of short-/medium-
chain-length poly(hydroxyalkanoate) blends by mixed culture
fermentation of glycerol. Biomacromolecules 2005;6:2106–12.
[140] Ashby RD, Shi FY, Gross RA. Use of poly(ethylene glycol)
to control the end group structure and molecular weight of
poly(3-hydroxybutyrate) formed by Alcaligenes latus DSM 1122.
Tetrahedron 1997;53:15209–23.
[141] Shi FY, Ashby R, Gross RA. Use of poly(ethylene glycol)s to regu-
late poly(3-hydroxybutyrate) molecular weight during Alcaligenes
eutrophus cultivations. Macromolecules 1996;29:7753–8.
[142] Sanguanchaipaiwong V, Gabelish CL, Hook J, Scholz C, Fos-
ter LJR. Biosynthesis of natural-synthetic hybrid copolymers:
polyhydroxyoctanoate-diethylene glycol. Biomacromolecules
2004;5:643–9.
[143] Saha SP, Patra A, Paul AK. Incorporation of polyethylene gly-
col in polyhydroxyalkanoic acids accumulated by Azotobacter
chroococcum MAL-201. Journal of Industrial Microbiology and
Biotechnology 2006;33:377–83.
[144] Zanzig J, Scholz C. Effects of poly(ethylene glycol) on the production
of poly(␤-hydroxybutyrate) by Azotobacter vinelandii UWD. Journal
of Polymers and the Environment 2003;11:145–54.
[145] Suzuki T, Deguchi H, Yamane T, Shimizu S, Gekko K. Control of
molecular-weight of poly-␤-hydroxybutyric acid produced in fet-
batch culture of Protomonas-Extorquens. Applied Microbiology
and Biotechnology 1988;27:487–91.
[146] Doi Y, Segawa A, Kawaguchi Y, Kunioka M. Cyclic nature of
poly(3-hydroxyalkanoate) metabolism in Alcaligenes-Eutrophus.
FEMS Microbiology Letters 1990;67:165–9.
[147] Taidi B, Mansfield DA, Anderson AJ. Turnover of poly(3-
hydroxybutyrate) (P(3HB)) and its influence on the molecular-mass
of the polymer accumulated by Alcaligenes-Eutrophus during batch
culture. FEMS Microbiology Letters 1995;129:201–5.
[148] Penloglou G, Roussos A, Chatzidoukas C, Kiparissides C. A
combined metabolic/polymerization kinetic model on the micro-
bial production of poly(3-hydroxybutyrate). New Biotechnology
2010;27:358–67.
[149] Barnard GN, Sanders JKM. Observation of mobile poly(␤-
hydroxybutyrate) in the storage granules of Methylobacterium Am1
by in vivo C-13-Nmr spectroscopy. FEBS Letters 1988;231:16–8.
[150] Barnard GN, Sanders JKM. The poly-␤-hydroxybutyrate granule
in vivo – a new insight based on NMR-spectroscopy of whole cells.
Journal of Biological Chemistry 1989;264:3286–91.
[151] Kawaguchi Y, Doi Y. Structure of native poly(3-hydroxybutyrate)
granules characterized by X-ray-diffraction. FEMS Microbiology
Letters 1990;70:151–6.
[152] Sanders JKM. NMR of nature’s plastics and spider’s webs – chem-
istry, physics, or biology. Chemical Society Reviews 1993;22:1–7.
[153] de Koning GJM, Lemstra PJ. The amorphous state of bac-
terial poly[(R)-3-hydroxyalkanoate] in vivo. Polymer 1992;33:
3292–4.
[154] Bonthrone KM, Clauss J, Horowitz DM, Hunter BK, Sanders
JKM. The biological and physical chemistry of polyhydroxyalka-
noates as seen by NMR-spectroscopy. FEMS Microbiology Review
1992;103:269–77.
[155] Barham PJ. Nucleation behavior of poly-3-hydroxy-butyrate. Jour-
nal of Materials Science 1984;19:3826–34.
[156] Organ SJ, Barham PJ. Nucleation, growth and morphology of
poly(hydroxybutyrate) and its copolymers. Journal of Materials Sci-
ence 1991;26:1368–74.
[157] Horowitz DM, Sanders JKM. Biomimetic, amorphous granules
of polyhydroxyalkanoates – composition, mobility, and stabi-
lization in vitro by proteins. Canadian Journal of Microbiology
1995;41(13):115–23.
[158] Lauzier C, Revol JF, Marchessault RH. Topotactic crystallization
of isolated poly(␤-hydroxybutyrate) granules from Alcaligenes
eutropha. FEMS Microbiology Review 1992;103:299–310.
[159] Harrison STL, Chase HA, Amor SR, Bonthrone KM, Saunders JKM.
Plasticization of poly(hydroxybutyrate) in vivo. International Jour-
nal of Biological Macromolecules 1992;14:50–6.
[160] Barham PJ, Keller A, Otun EL, Holmes PA. Crystallization and mor-
phology of a bacterial thermoplastic – poly-3-hydroxybutyrate.
Journal of Materials Science 1984;19:2781–94.
[161] Tanaka T, Fujita M, Takeuchi A, Suzuki Y, Uesugi K, Doi Y. Structure
investigation of narrow banded spherulites in polyhydroxyalka-
noates by microbeam X-ray diffraction with synchrotron radiation.
Polymer 2005;46:5673–9.
[162] Fujita M, Sawayanagi T, Tanaka T, Iwata T, Abe H, Doi Y. Synchrotron
SAXS and WAXS studies on changes in structural and thermal prop-
erties of poly[(R)3-hydroxybutyrate] single crystals during heating.
Macromolecular Rapid Communications 2005;26:678–83.
[163] Yokouchi M, Chatani Y, Tadokoro H, Teranish K, Tani H. Struc-
tural studies of polyesters. 5. Molecular and crystal structures
of optically-active and racemic poly(␤-hydroxybutyrate). Polymer
1973;14:267–72.
[164] Cornibert J, Marchessault RH. Physical properties of poly-␤-
hydroxybutyrate. IV. Conformational analysis and crystalline
structure. Journal of Molecular Biology 1972;71:735–56.
[165] Scandola M, Ceccorulli G, Pizzoli M, Gazzano M. Study of the crystal
phase and crystallization rate of bacterial poly(3-hydroxybutyrate-
co-3-hydroxyvalerate). Macromolecules 1992;25:1405–10.
[166] Marchessault RH, Morikawa H, Revol JF, Bluhm TL. Physical
properties of a naturally-occurring polyester: poly(␤-
hydroxyvalerate)/poly(␤-hydroxybutyrate). Macromolecules
1984;17:1882–4.
[167] Iwata T, Doi Y. Crystal structure and biodegradation of aliphatic
polyester crystals. Macromolecular Chemistry and Physics
1999;200:2429–42.
[168] Marchessault RH, Debzi EM, Revol JF, Steinbüchel A. Single-crystals
of bacterial and synthetic poly(3-hydroxyvalerate). Canadian Jour-
nal of Microbiology 1995;41(13):297–302.
[169] Marchessault RH, Coulombe S, Morikawa H, Okamura K, Revol
JF. Solid-state properties of poly-␤-hydroxybutyrate and of its
oligomers. Canadian Journal of Chemistry 1981;59:38–44.
[170] Sato H, Murakami R, Mori K, Ando Y, Takahashi I, Noda I. Spe-
cific crystal structure of poly(3-hydroxybutyrate) thin films studied
by infrared reflection-absorption spectroscopy. Vibrational Spec-
troscopy 2009;51:132–5.
[171] Lundgren DG, Alper R, Schnaitman C, Marchessault RH. Char-
acterization of poly-␤-hydroxybutyrate extracted from different
bacteria. Journal of Bacteriology 1965;89:245–51.
[172] Mitomo H, Barham PJ, Keller A. Crystallization and morphology
of poly(␤-hydroxybutyrate) and its copolymer. Polymer Journal
1987;19:1241–53.
[173] Welland EL, Stejny J, Halter A, Keller A. Selective degradation of
chain folded single-crystals of poly(␤-hydroxybutyrate). Polymer
Communications 1989;30:302–4.
[174] Birley C, Briddon J, Sykes KE, Barker PA, Organ SJ, Barham
PJ. Morphology of single-crystals of poly(hydroxybutyrate) and
copolymers of hydroxybutyrate and hydroxyvalerate. Journal of
Materials Science 1995;30:633–8.
[175] Revol JF, Chanzy HD, Deslandes Y, Marchessault RH. High-
resolution electron-microscopy of poly(␤-hydroxybutyrate). Poly-
mer 1989;30:1973–6.
[176] Nobes GAR, Briese BH, Jendrossek D, Marchessault RH. Microscopic
visualization of the enzymatic degradation of poly(3HB-co-3HV)
and poly(3HV) single crystals by PHA depolymerases from Pseu-
domonas lemoignei. Journal of Environmental Polymer Degradation
1998;6:99–107.
[177] Marchessault RH, Dou HY, Ramsay J. Microbial medium chainlength
poly[(R)-3-hydroxyalkanoate] shows liquid crystal behaviour.
International Journal of Biological Macromolecules 2011;48:
271–5.
[178] Lambeek G, Vorenkamp EJ, Schouten AJ. Structural study
of Langmuir–Blodgett monolayers and multilayers of poly(␤-
hydroxybutyrate). Macromolecules 1995;28:2023–32.
[179] Orts WJ, Marchessault RH, Bluhm TL, Hamer GK. Observation
of strain-induced ␤-form in poly(␤-hydroxyalkanoates). Macro-
molecules 1990;23:5368–70.
[180] Yamane H, Terao K, Hiki S, Kimura Y. Mechanical properties and
higher order structure of bacterial homo poly(3-hydroxybutyrate)
melt spun fibers. Polymer 2001;42:3241–8.
[181] Fischer JJ, Aoyagi Y, Enoki M, Doi Y, Iwata T. Mechanical properties
and enzymatic degradation of poly([R]-3-hydroxybutyrate-co-
[R]-3-hydroxyhexanoate) uniaxially cold-drawn films. Polymer
Degradation and Stability 2004;83:453–60.
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
[182] Tanaka T, Fujita M, Takeuchi A, Suzuki Y, Uesugi K, Ito K. Forma-
tion of highly ordered structure in poly[(R)-3-hydroxybutyrate-
co-(R)-3-hydroxyvalerate] high-strength fibers. Macromolecules
2006;39:2940–6.
[183] Iwata T, Sato S, Park JW. ␤ structure and unique crystalline ori-
entation analysis of P(3HB) fibers and films. In: 11th international
symposium on biological polyesters. 2008. p. 101.
[184] Kusaka S, Iwata T, Doi Y. Properties and biodegradability of ultra-
high-molecular-weight poly[(R)-3-hydroxybutyrate] produced by
a recombinant Escherichia coli. International Journal of Biological
Macromolecules 1999;25:87–94.
[185] Furuhashi Y, Imamura Y, Jikihara Y, Yamane H. Higher order
structures and mechanical properties of bacterial homo poly(3-
hydroxybutyrate) fibers prepared by cold-drawing and annealing
processes. Polymer 2004;45:5703–12.
[186] Nishiyama Y, Tanaka T, Yamazaki T, Iwata T. 2D NMR observation
of strain-induced ␤-form in poly[(R)-3-hydroxybutyrate]. Macro-
molecules 2006;39:4086–92.
[187] Murakami R, Dybal J, Iwata T, Ozaki Y, Sato H. Formation and sta-
bility of ␤-structure in biodegradable ultra-high-molecular-weight
poly(3-hydroxybutyrate) by infrared, Raman, and quantum chem-
ical calculation studies. Polymer 2007;48:2672–80.
[188] Ublekov F, Baldrian J, Nedkov E. Crystalline ␤-structure of P(3HB)V
grown epitaxially on silicate layers of MMT. Journal of Polymer
Science Part B: Polymer Physics 2009;47:751–5.
[189] Antipov EM, Dubinsky VA, Rebrov AV, Nekrasov YP, Gordeev
SA, Ungar G. Strain-induced mesophase and hard-elastic
behaviour of biodegradable polyhydroxyalkanoates fibers.
Polymer 2006;47:5678–90.
[190] Bluhm TL, Hamer GK, Marchessault RH, Fyfe CA, Veregin
RP. Isodimorphism in bacterial poly(␤-hydroxybutyrate-co-␤-
hydroxyvalerate). Macromolecules 1986;19:2871–6.
[191] Inoue Y. Solid-state structure and properties of bacterial
copolyesters. Journal of Molecular Structure 1998;441:119–27.
[192] Kunioka M, Tamaki A, Doi Y. Crystalline and
thermal properties of bacterial copolyesters – of poly(3-hydroxybutyrate) and poly(3-hydroxybutyrate-co-3-
poly(3-hydroxybutyrate-co-3-hydroxyvalerate) and poly(3-
hydroxybutyrate-co-4-hydroxybutyrate). Macromolecules
1989;22:694–7.
[193] Mitomo H, Morishita N, Doi Y. Composition range of crystal
phase-transition of isodimorphism in poly(3-hydroxybutyrate-co-
3-hydroxyvalerate). Macromolecules 1993;26:5809–11.
[194] Mitomo H, Morishita N, Doi Y. Structural changes of poly(3-
hydroxybutyrate-co-3-hydroxyvalerate) fractionated with
acetone–water solution. Polymer 1995;36:2573–8.
[195] Mitomo H, Doi Y. Lamellar thickening and cocrystallization of
poly(hydroxyalkanoate)s on annealing. International Journal of
Biological Macromolecules 1999;25:201–5.
[196] Ishida K, Asakawa N, Inoue Y. Structure, properties and biodegrada-
tion of some bacterial copoly(hydroxyalkanoate)s. Macromolecular
Symposium 2005;224:47–57.
[197] Feng LD, Yoshie Y, Asakawa N, Inoue Y. Comonomer-unit composi-
tions, physical properties and biodegradability of bacterial copoly-
hydroxyalkanoates. Macromolecular Bioscience 2004;4:186–98.
[198] Feng LD, Wang Y, Inagawa Y, Kasuya K, Saito T, Doi Y. Enzymatic
degradation behavior of comonomer compositionally fraction-
ated bacterial poly(3-hydroxybutyrate-co-3-hydroxyvalerate)s by
poly(3-hydroxyalkanoate) depolymerases isolated from Ralstonia
pickettii T1 and Acidovorax sp TP4. Polymer Degradation and Sta-
bility 2004;84:95–104.
[199] Tsz-Chun M, Chan PL, Lawford H, Chua H, Lo WH, Yu PH. Micro-
bial synthesis and characterization of physiochemical properties
of polyhydroxyalkanoates (PHAs) produced by bacteria isolated
from activated sludge obtained from the municipal wastewater
works in Hong Kong. Applied Biochemistry and Biotechnology
2005;121:731–9.
[200] Thellen C, Coyne M, Froio D, Auerbach M, Wirsen C, Ratto JA. A
processing, characterization and marine biodegradation study of
melt-extruded polyhydroxyalkanoate (PHA) films. Journal of Poly-
mers and the Environment 2008;16:1–11.
[201] Wang Y, Yamada S, Asakawa N, Yamane T, Yoshie N, Inoue Y.
Comonomer compositional distribution and thermal and mor-
phological characteristics of bacterial poly(3-hydroxybutyrate-
co-3-hydroxyvalerate)s with high 3-hydroxyvalerate content.
Biomacromolecules 2001;2:1315–23.
[202] Yamada S, Wang Y, Asakawa N, Yoshie N, Inoue Y. Crys-
talline structural change of bacterial poly(3-hydroxybutyrate-
co-3-hydroxyvalerate) with narrow compositional distribution.
Macromolecules 2001;34:4659–61.
437
[203] Mitomo H, Takahashi T, Ito H, Saito T. Biosynthesis and charac-
terization of poly(3-hydroxybutyrate-co-3-hydroxyvalerate) pro-
duced by Burkholderia cepacia D1. International Journal of Biological
Macromolecules 1999;24:311–8.
[204] Iwata T, Doi Y, Nakayama S, Sasatsuki H, Teramachi S. Structure
and enzymatic degradation of poly(3-hydroxybutyrate) copolymer
single crystals with an extracellular P(3HB) depolymerase from
Alcaligenes faecalis T1. International Journal of Biological Macro-
molecules 1999;25:169–76.
[205] Kamiya N, Sakurai M, Inoue Y, Chujo R, Doi Y. Studies of cocrys-
tallization of poly(3-hydroxybutyrate-co-3-hydroxyvalerate)
by solid-state high-resolution C-13 NMR-spectroscopy and
differential scanning calorimetry. Macromolecules 1991;24:
2178–82.
[206] Yoshie N, Fujiwara M, Kasuya K, Abe H, Doi Y, Inoue Y. Effect of
monomer composition and composition distribution on enzymatic
degradation of poly(3-hydroxybutyrate-co-3-hydroxyvalerate).
Macromolecular Chemistry and Physics 1999;200:977–82.
[207] Barker PA, Mason F, Barham PJ. Density and crystallinity of
poly(3-hydroxybutyrate 3-hydroxyvalerate) copolymers. Journal
of Materials Science 1990;25:1952–6.
[208] Yoshie N, Saito M, Inoue Y. Structural transition of lamella
crystals in a isomorphous copolymer, poly(3-hydroxybutyrate-co-
3-hydroxyvalerate). Macromolecules 2001;34:8953–60.
[209] Barker PA, Barham PJ, Martinez-Salazar J. Effect of crystal-
lization temperature on the cocrystallization of hydroxybu-
tyrate/hydroxyvalerate copolymers. Polymer 1997;38:913–9.
[210] Yoshie N, Sakurai M, Inoue Y, Chujo R. Cocrystallization of isother-
mally crystallized poly(3-hydroxybutyrate-co-3-hydroxyvalerate).
Macromolecules 1992;25:2046–8.
[211] Barham PJ, Barker P, Organ SJ. Physical properties of
poly(hydroxybutyrate) and copolymers of hydroxybutyrate and
hydroxyvalerate. FEMS Microbiology Review 1992;103:289–98.
[212] Yoshie N, Fujiwara M, Ohmori M, Inoue Y. Temperature depend-
ence of cocrystallization and phase segregation in blends
hydroxyvalerate). Polymer 2001;42:8557–63.
[213] Guo LH, Sato H, Hashimoto T, Ozaki Y. Thermally induced exchanges
of hydrogen bonding interactions and their effects on phase struc-
tures of poly(3-hydroxybutyrate) and poly(4-vinylphenol) blends.
Macromolecules 2011;44:2229–39.
[214] Chen Y, Yang G, Chen Q. Solid-state NMR study on the structure and
mobility of the noncrystalline region of poly(3-hydroxybutyrate)
and poly(3-hydroxybutyrate-co-3-hydroxyvalerate). Polymer
2002;43:2095–9.
[215] Orts WJ, Vanderhart DL, Bluhm TL, Marchessault RH. Cocrys-
tallization in random copolymers of poly(␤-hydroxybutyrate-
co-␤-hydroxyvalerate) and its effect on crystalline morphology.
Canadian Journal of Chemistry 1995;73:2094–100.
[216] Xie YP, Noda I, Akpaiu YA. Influence of cooling rate on the thermal
behavior and solid-state morphologies of polyhydroxyalkanoates.
Journal of Applied Polymer Science 2008;109:2259–68.
[217] Cao A, Asakawa N, Yoshie N, Inoue Y. High-resolution solid-state
C-13 n. m. r. study on phase structure of the compositionally
fractionated bacterial copolyester poly(3-hydroxybutyric acid-co-
3-hydroxypropionic acid)s. Polymer 1999;40:3309–22.
[218] Runt J, Kanchanasopa M. Crystallinity determination. Encyclopedia
of polymer science and technology, vol. 9, 4th ed. New York: John
Wiley & Sons Inc.; 2004. p. 446–64.
[219] Zhang LM, Tang HR, Hou GJ, Shen YD, Deng F. The domain struc-
ture and mobility of semi-crystalline poly(3-hydroxybutyrate) and
poly(3-hydroxybutyrate-co-3-hydroxyvalerate): a solid-state NMR
study. Polymer 2007;48:2928–38.
[220] Kansiz M, Dominguez-Vidal A, McNaughton D, Lendl B. Fourier-
transform infrared (FTIR) spectroscopy for monitoring and
determining the degree of crystallisation of polyhydroxyalka-
noates (PHAs). Analytical and Bioanalytical Chemistry 2007;388:
1207–13.
[221] Bauer H, Owen AJ. Some structural and mechanical prop-
erties of bacterially produced poly-␤-hydroxybutyrate-co-␤-
hydroxyvalerate. Colloid and Polymer Science 1988;266:241–7.
[222] Steinbüchel A, Schmack G. Large-scale production of poly(3-
hydroxyvaleric acid) by fermentation of Chromobacterium vio-
laceleum, processing, and characterization of the homopolyester.
Journal of Environmental Polymer Degradation 1995;3:243–58.
[223] Pearce RP, Marchessault RH. Melting and crystallization in
bacterial poly(␤-hydroxyvalerate), PHV, and blends with
poly(␤-vhydroxybutyrate-co-hydroxyvalerate). Macromolecules
1994;27:3869–74.
438 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
[224] Bloembergen S, Holden DA, Marchessault RH. Non-biochemical
synthesis and characterization of poly(␤-hydroxybutyrate-co-␤-
hydroxyvalerate). Polymer Preprints (American Chemical Society,
Division of Polymer Chemistry) 1988;29(1):594–5.
[225] Gassner F, Owen AJ. Some properties of poly(3-
hydroxybutyrate)–poly(3-hydroxyvalerate) blends. Polymer
International 1996;39:215–9.
[226] Cheng ML, Sun YM. Relationship between free volume properties
and structure of poly(3-hydroxybutyrate-co-3-hydroxyvalerate)
membranes via various crystallization conditions. Polymer
2009;50:5298–307.
[227] Chan CH, Kummerlowe C, Kammer HW. Crystallization and melting
behavior of poly(3-hydroxybutyrate)-based blends. Macromolecu-
lar Chemistry and Physics 2004;205:664–75.
[228] Orts WJ, Romansky M, Guillet JE. Measurement of the crystallinity
of poly(␤-hydroxybutyrate-co-␤-hydroxyvalerate) copolymers
by inverse gas-chromatography. Macromolecules 1992;25:
949–53.
[229] Kamiya N, Sakurai M, Inoue Y, Chujo R. Isomorphic behavior of ran-
dom copolymers – thermodynamic analysis of co-crystallization of
poly(3-hydroxybutyrate-co-3-hydroxyvalerate). Macromolecules
1991;24:3888–92.
[230] Chen HL, Hwang JC. Some comments on the degree of crys-
tallinity defined by the enthalpy of melting. Polymer 1995;36:
4355–7.
[231] Sawayanagi T, Tanaka T, Iwata T, Abe H, Doi Y, Ito K, Fujisawa
T, Fujita M. Real-time synchrotron SAXS and WAXD studies on
annealing behavior of poly[(R)-3-hydroxybutyrate] single crystals.
Macromolecules 2006;39:2201–8.
[232] Poley LH, Siqueira APL, da Silva MG, Sanchez R, Prioli R, Mansanares
AM, Vargas H. Photothermal methods and atomic force microscopy
images applied to the study of poly(3-hydroxybutyrate)
and poly(3-hydroxybutyrate-co-3-hydroxyvalerate) dense
membranes. Journal of Applied Polymer Science 2005;97:
1491–7.
[233] Lotti N, Pizzoli M, Ceccorulli G, Scandola M. Binary blends of
microbial poly(3-hydroxybutyrate) with polymethacrylates. Poly-
mer 1993;34:4935–40.
[234] Gunaratne LMWK, Shanks RA. Multiple melting behaviour of
poly(3-hydroxybutyrate-co-hydroxyvalerate) using step-scan DSC.
European Polymer Journal 2005;41:2980–8.
[235] Liu T, Petermann J. Multiple melting behavior in isothermally cold-
crystallized isotactic polystyrene. Polymer 2001;42:6453–61.
[236] Gan ZH, Abe H, Doi Y. Biodegradable poly(ethylene succinate) (PES).
1. Crystal growth kinetics and morphology. Biomacromolecules
2000;1:704–12.
[237] Scandola M, Pizzoli M, Ceccorulli G, Cesaro A, Paoletti S,
Navarini L. Viscoelastic and thermal properties of bacterial
poly(d-(−)-␤-hydroxybutyrate). International Journal of Biological
Macromolecules 1988;10:373–7.
[238] Hoffman JD, Miller RL. Kinetic of crystallization from the melt and
chain folding in polyethylene fractions revisited: theory and exper-
iment. Polymer 1997;38:3151–212.
[239] Marand H, Xu JN, Srinivas S. Determination of the equilib-
rium melting temperature of polymer crystals: linear and
nonlinear Hoffman–Weeks extrapolations. Macromolecules
1998;31:8219–29.
[240] Organ SJ. Variation in melting point with molecular weight
for hydroxybutyrate hydroxyvalerate copolymers. Polymer
1993;34:2175–9.
[241] Organ SJ, Barham PJ. On the equilibrium melting temperature of
polyhydroxybutyrate. Polymer 1993;34:2169–74.
[242] Kemnitzer JE, Gross RA, Mccarthy SP, Liggat J, Blundell DJ, Cox
M. Crystallization behavior of predominantly syndiotactic poly(␤-
hydroxybutyrate). Journal of Environmental Polymer Degradation
1995;3:37–47.
[243] Bloembergen S, Holden DA, Hamer GK, Bluhm TL, Marches-
sault RH. Studies of composition and crystallinity of bacterial
poly(␤-hydroxybutyrate-co-␤-hydroxyvalerate). Macromolecules
1986;19:2865–71.
[244] Furukawa T, Sato H, Murakami R, Zhang JM, Duan YX, Noda I,
Ochiai S, Ozaki Y. Structure, dispersibility, and crystallinity of
poly(hydroxybutyrate)/poly(l-lactic acid) blends studied by FT-IR
microspectroscopy and differential scanning calorimetry. Macro-
molecules 2005;38:6445–54.
[245] Sato H, Murakami R, Padermshoke A, Hirose F, Senda K, Noda I,
Ozaki Y. Infrared spectroscopy studies of CH••O hydrogen bondings
and thermal behavior of biodegradable poly(hydroxyalkanoate).
Macromolecules 2004;37:7203–13.
[246] Wen X, Lu X, Peng Q, Zhu F, Zheng N. Crystallization behaviors
and morphology of biodegradable poly(3-hydroxybutyrate-co-4-
hydroxybutyrate). Journal of Thermal Analysis and Calorimetry
2011, http://dx.doi.org/10.1007/s10973-011-1768-2.
[247] Gagnon KD, Lenz RW, Farris RJ, Fuller RC. Crystallization behavior
and its influence on the mechanical properties of a thermoplastic
elastomer produced by Pseudomonas-Oleovorans. Macromolecules
1992;25:3723–8.
[248] de Koning GJM, van Bilsen HMM, Lemstra PJ, Hazenberg W, With-
olt B, Preusting H, van der Galiën JG, Schirmer A, Jendrossek D.
A biodegradable rubber by cross-linking poly(hydroxyalkanoate)
from Pseudomonas oleovorans. Polymer 1994;35:2090–7.
[249] Marchessault RH, Monasterios CJ, Morin FG, Sundararajan PR. Chi-
ral poly(␤-hydroxyalkanoates) – an adaptable helix influenced by
the alkane side-chain. International Journal of Biological Macro-
molecules 1990;12:158–65.
[250] Zhao K, Deng Y, Chen GQ. Effects of surface morphology on the bio-
compatibility of polyhydroxyalkanoates. Biochemical Engineering
Journal 2003;16:115–23.
[251] Chanprateep S, Kikuya K, Shimizu H, Seki S, Tagawa S,
Shioya S. Nonisothermal crystallization kinetics of biodegrad-
able random poly(3-hydroxybutyrate-co-3-hydroxyvalerate) and
block one. Journal of Chemical Engineering of Japan 2003;36:
639–46.
[252] Chuah KP, Gan SN, Chee KK. Determination of Avrami exponent by
differential scanning calorimetry for non-isothermal crystallization
of polymers. Polymer 1999;40:253–9.
[253] Shan GF, Gong X, Chen WP, Chen L, Zhu MF. Effect of
multi-walled carbon nanotubes on crystallization behavior of
poly(3-hydroxybutyrate-co-3-hydroxyvalerate). Colloid and Poly-
mer Science 2011;289:1005–14.
[254] Peng SW, An YX, Chen C, Fei B, Zhuang YG, Dong LS. Isothermal
crystallization of poly(3-hydroxybutyrate-co-3-hydroxyvalerate).
European Polymer Journal 2003;39:1475–80.
[255] Liu WJ, Yang HL, Wang Z, Dong LS, Liu JJ. Effect of nucleating
agents on the crystallization of poly(3-hydroxybutyrate-
co-3-hydroxyvalerate). Journal of Applied Polymer Science
2002;86:2145–52.
[256] Saad GR, Mansour AA, Hamed AH. Dielectric investigation
of cold crystallization of poly(3-hydroxybutyrate). Polymer
1997;38:4091–6.
[257] Xu J, Guo BH, Yang R, Wu Q, Chen GQ, Zhang ZM. In situ FTIR study
on melting and crystallization of polyhydroxyalkanoates. Polymer
2002;43:6893–9.
[258] Ziaee Z, Supaphol P. Non-isothermal melt- and cold-
crystallization kinetics of poly(3-hydroxybutyrate). Polymer
Testing 2006;25:807–18.
[259] Kammer HW, Kressler J, Kummerloewe C. Phase-behavior of poly-
mer blends – effects of thermodynamics and rheology. Advances in
Polymer Science 1993;106:31–85.
[260] Hill DJ, Markotsis M, Whittaker AK, Wong KW. NMR characteriza-
tion of blends of poly(hydroxybutyrate-co-hydroxyvalerate) with
poly(vinyl acetate). Polymer International 2003;52:1780–9.
[261] Olabisi O, Robeson LM, Shaw MT. Polymer–polymer miscibility.
New York: Academic Press; 1979, 370 pp.
[262] An YX, Li G, Mo ZS, Feng ZL, Dong LS. Miscibility, crystalliza-
tion kinetics, and morphology of poly(␤-hydroxybutyrate) and
poly(methyl acrylate) blends. Journal of Polymer Science Part B:
Polymer Physics 2000;38:1860–7.
[263] Cox MK. Recycling biopol – composting and material recycling.
Journal of Macromolecular Science: Pure and Applied Chemistry
1995;A32:607–12.
[264] Luzier WD. Materials derived from biomass biodegradable materi-
als. Proceedings of the National Academy of Sciences of the United
States of America 1992;89:839–42.
[265] Anderson AJ, Williams DR, Taidi B, Dawes EA, Ewing DF. Studies
on copolyester synthesis by Rhodococcus ruber and factors influ-
encing the molecular mass of polyhydroxybutyrate accumulated
by Methylobacterium extorquens and Alcaligenes-eutrophus. FEMS
Microbiology Review 1992;103:93–101.
[266] Van der Walle GAM, de Koning GJM, Weusthuis RA, Eggink G. Prop-
erties, modifications and applications of biotechnology. Advances
in Biochemical Engineering/Biotechnology 2001;71:264–91.
[267] van Wegen RJ, Lee SY, Middelberg APJ. Metabolic and kinetic
analysis of poly(3-hydroxybutyrate) production by recombinant
Escherichia coli. Biotechnology and Bioengineering 2001;74:70–80.
[268] Reusch RN. Low-molecular-weight complexed poly(3-
hydroxybutyrate) – a dynamic and versatile molecule in vivo.
Canadian Journal of Microbiology 1995;41(13):50–4.
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
[269] Fiorese ML, Freitas F, Pais J, Ramos AM, de Araga GMF, Reis MAM.
Recovery of polyhydroxybutyrate (P(3HB)) from Cupriavidus neca-
tor biomass by solvent extraction with 1,2-propylene carbonate.
Engineering in Life Sciences 2009;9:454–61.
[270] Akita S, Einaga Y, Miyaki Y, Fujita H. Solution properties of
poly(d-␤-hydroxybutyrate). 1. Biosynthesis and characterization.
Macromolecules 1976;9:774–80.
[271] Dias JML, Serafim LS, Lemos PC, Reis MAM, Oliveira R. Mathematical
modelling of a mixed culture cultivation process for the produc-
tion of polyhydroxybutyrate. Biotechnology and Bioengineering
2005;92:209–22.
[272] Dias JML, Oehmen A, Serafim LS, Lemos PC, Reis MAM, Oliveira
R. Metabolic modelling of polyhydroxyalkanoate copolymers pro-
duction by mixed microbial cultures. BMC Systems Biology
2008;2:59–80.
[273] Bradel R, Reichert KH. Modeling of molar-mass distribution
of poly(d-(−)-3-hydroxybutyrate) during bacterial synthesis. Die
Makromolekulare Chemie 1993;194:1983–90.
[274] Jurasek L, Marchessault RH. Polyhydroxyalkanoate (PHA) gran-
ule formation in Ralstonia eutropha cells: a computer simulation.
Applied Microbiology and Biotechnology 2004;64:611–7.
[275] Mantzaris NV, Kelley AS, Daoutidis P, Srienc F. A population balance
model describing the dynamics of molecular weight distributions
and the structure of PHA copolymer chains. Chemical Engineering
Science 2002;57:4643–63.
[276] Iadevaia S, Mantzaris NV. Genetic network driven control
of P(3HB)V copolymer composition. Journal of Biotechnology
2006;122:99–121.
[277] Yamanaka K, Aoki T, Kudo T, Kimura Y. Effect of ethylene glycol
on the end group structure of poly(3-hydroxybutyrate). Polymer
Degradation and Stability 2010;95:1284–91.
[278] Koizumi F, Abe H, Doi Y. Molecular-weight of poly(3-
hydroxybutyrate) during biological polymerization in
Alcaligenes-eutrophus. Journal of Macromolecular Science:
Pure and Applied Chemistry 1995;A32:759–74.
[279] Shimizu H, Tamura S, Shioya S, Suga K. Kinetic study of poly-d(−)-3-
hydroxybutyric acid (P(3HB)) production and its molecular-weight
distribution control in a fed-batch culture of Alcaligenes eutrophus.
Journal of Fermentation and Bioengineering 1993;76:465–9.
[280] Hiraishi T, Kikkawa Y, Fujita M, Normi YM, Kanesato M, Tsuge
T, Sudesh K, Maeda M, Doi Y. Atomic force microscopic observa-
tion of in vitro polymerized poly[(R)-3-hydroxybutyrate]: insight
into possible mechanism of granule formation. Biomacromolecules
2005;6:2671–7.
[281] Kurja J, Zirkzee HF, de Koning GM, Maxwell IA. A new
kinetic-model for the accumulation of poly-3-hydroxybutyrate in
Alcaligenes-eutrophus, 1. Granule growth. Macromolecular Theory
and Simulations 1995;4:839–55.
[282] Sim SJ, Snell KD, Hogan SA, Stubbe J, Rha CK, Sinskey AJ. PHA syn-
thase activity controls the molecular weight and polydispersity of
polyhydroxybutyrate in vivo. Nature Biotechnology 1997;15:63–7.
[283] Huisman GW, Wonink E, de Koning G, Preusting H, Witholt B.
Synthesis of poly(3-hydroxyalkanoates) by mutant and recombi-
nant Pseudomonas strains. Applied Microbiology and Biotechnology
1992;38:1–5.
[284] Harris JM, editor. Poly(ethylene glycol) chemistry – biotechnical
and biomedical applications. New York: Plenum Press; 1992, 408
pp.
[285] Yamazaki M, Ito T. Deformation and instability in membrane-
structure of phospholipid-vesicles caused by osmophobic asso-
ciation – mechanical-stress model for the mechanism of
poly(ethylene glycol)-induced membrane-fusion. Biochemistry
1990;29:1309–14.
[286] Helm J, Wendlandt KD, Jechorek M, Stottmeister U. Potassium
deficiency results in accumulation of ultra-high molecular weight
poly-␤-hydroxybutyrate in a methane-utilizing mixed culture.
Journal of Applied Microbiology 2008;105:1054–61.
[287] Taidi B, Anderson AJ, Dawes EA, Byrom D. Effect of carbon
source and concentration on the molecular-mass of poly(3-
hydroxybutyrate) produced by Methylobacterium-extorquens and
Alcaligenes-eutrophus. Applied Microbiology and Biotechnology
1994;40:786–90.
[288] McChalicher CWJ, Srienc F. Investigating the structure–property
relationship of bacterial PHA block copolymers. Journal of Biotech-
nology 2007;132:296–302.
[289] Curley JM, Lenz RW, Fuller RC. Sequential production of two
different polyesters in the inclusion bodies of Pseudomonas oleovo-
rans. International Journal of Biological Macromolecules 1996;19:
29–34.
439
[290] Kelley AS, Srienc F. Production of two phase polyhydroxyalkanoic
acid granules in Ralstonia eutropha. International Journal of Biolog-
ical Macromolecules 1999;25:61–7.
[291] Yoon SC, Choi MH. Local sequence dependence of polyhydroxyalka-
noic acid degradation in Hydrogenophaga pseudoflava. Journal of
Biological Chemistry 1999;274:37800–8.
[292] Pederson EN, Srienc F. Mass spectrometry feedback control for
synthesis of polyhydroxyalkanoate granule microstructures in Ral-
stonia eutropha. Macromolecular Bioscience 2004;4:243–54.
[293] Madden LA, Asrar J, Anderson AJ. Synthesis and characterization
of poly(3-hydroxybutyrate) and poly(3-hydroxybutyrate-co-3-
hydroxyvalerate) polymer mixtures produced in high-density
fed-batch cultures of Ralstonia eutropha (Alcaligenes eutrophus).
Macromolecules 1998;31:5660–7.
[294] Mantzaris NV, Kelley AS, Srienc F, Daoutidis P. Optimal carbon
source switching strategy for the production of PHA copolymers.
AIChE Journal 2001;47:727–43.
[295] Kelley AS, Srienc F. Controlling the polymer microstructure of
biodegradable polyhydroxyalkanoates. In: Gross RA, Cheng HN,
editors. Biocatalysis in polymer science. ACS Symposium Series, vol.
840. Washington, DC: American Chemical Society; 2002. p. 124–7.
[296] Chanprateep S, Shimizu H, Shioya S. Characterization and enzy-
matic degradation of microbial copolyester P(3HB-co-3HV)s
produced by metabolic reaction model-based system. Polymer
Degradation and Stability 2006;91:2941–50.
[297] Pederson EN, McChalicher CWJ, Srienc F. Bacterial synthesis of PHA
block copolymers. Biomacromolecules 2006;7:1904–11.
[298] Li SY, Dong CL, Wang SY, Ye HM, Chen GQ. Microbial produc-
tion of polyhydroxyalkanoate block copolymer by recombinant
Pseudomonas putida. Applied Microbiology and Biotechnology
2011;90:659–69.
[299] Pereira SMF, Sanchez RJ, Rieumont J, Cabrera JG. Synthesis of
biodegradable polyhydroxyalcanoate copolymer from a renew-
able source by alternate feeding. Polymer Engineering and Science
2008;48:2051–9.
[300] Inoue Y, Yoshie N. Structure and physical properties of bac-
terially synthesized polyesters. Progress in Polymer Science
1992;17:571–610.
[301] Yoshie N, Saito M, Inoue Y. Effect of chemical composi-
tional distribution on solid-state structures and properties
of poly(3-hydroxybutyrate-co-3-hydroxyvalerate). Polymer
2004;45:1903–11.
[302] Yoshie N, Inoue Y. Chemical composition distribution of bacterial
copolyesters. International Journal of Biological Macromolecules
1999;25:193–200.
[303] Žagar E, Kržan A. Sequence distribution dependence on molar
mass in microbial poly(3-hydroxybutyrate-co-3-hydroxyvalerate)
copolyesters. Acta Chimica Slovenica 2009;56:386–91.
[304] Tanadchangsaeng N, Tsuge T, Abe H. Comonomer compo-
sitional distribution, physical properties, and enzymatic
degradability of bacterial poly(3-hydroxybutyrate-co-3-
hydroxy-4-methylvalerate) copolyesters. Biomacromolecules
2010;11:1615–22.
[305] Kamiya N, Yamamoto Y, Inoue Y, Chujo R, Doi Y. Micro-
structure of bacterially synthesized poly(3-hydroxybutyrate-co-3-
hydroxyvalerate). Macromolecules 1989;22:1676–82.
[306] Doi Y, Kunioka M, Nakamura Y, Soga K. Nuclear-magnetic-
resonance studies on poly(␤-hydroxybutyrate) and a copolyester
of ␤-hydroxybutyrate and ␤-hydroxyvalerate isolated from Alca-
ligenes eutrophus H16. Macromolecules 1986;19:2860–4.
[307] Inoue Y, Kamiya N, Yamamoto Y, Chujo R, Doi Y. Micro-
structures of commercially available poly(3-hydroxybutyrate-co-
3-hydroxyvalerate)s. Macromolecules 1989;22:3800–2.
[308] Adamus G, Sikorska W, Kowalczuk M, Montaudo M, Scandola
M. Sequence distribution and fragmentation studies of bacte-
rial copolyester macromolecules: characterization of P(3HB)V
macroinitiator by electrospray ion-trap multistage mass spectrom-
etry. Macromolecules 2000;33:5797–802.
[309] Ballistreri A, Montaudo G, Garozzo D, Chujo R, Doi Y. Microstruc-
ture of bacterial poly(␤-hydroxybutyrate-co-␤-hydroxyvalerate)
by fast-atom-bombardment mass-spectrometry: analysis of the
partial pyrolysis products. Macromolecules 1991;24:1231–6.
[310] Žagar E, Kržan A, Adamus G, Kowalczuk M. Sequence distribu-
tion in microbial poly(3-hydroxybutyrate-co-3-hydroxyvalerate)
co-polyesters determined by NMR and MS. Biomacromolecules
2006;7:2210–6.
[311] Lau NS, Tsuge T, Sudesh K. Formation of new polyhydroxyalkanoate
containing 3-hydroxy-4-methylvalerate monomer in Burkholderia
sp. Applied Microbiology and Biotechnology 2011;89:1599–609.
440 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
[312] Don TM, Chen CW, Chan TH. Preparation and characterization
of poly(hydroxyalkanoate) from the fermentation of Haloferax
mediterranei. Journal of Biomaterials Science, Polymer Edition
2006;17:1425–38.
[313] Yoshie N, Menju H, Sato H, Inoue Y. Complex composition distri-
bution of poly(3-hydroxybutyrate-co-3-hydroxyvalerate). Macro-
molecules 1995;28:6516–21.
[314] Inoue Y, Sano F, Nakamura K, Yoshie N, Saito Y, Satoh H.
Microstructure of copoly(3-hydroxyalkanoates) produced in the
anaerobic–aerobic activated sludge process. Polymer International
1996;39:183–9.
[315] Cao A, Ichikawa M, Kasuya KI, Yoshie N, Asakawa N, Inoue Y, Doi Y,
Abe H. Composition fractionation and thermal characterization of
poly(3-hydroxybutyrate-co-3-hydroxypropionate). Polymer Jour-
nal 1996;28:1096–102.
[316] Cao A, Ichikawa M, Ikejima T, Yoshie N, Inoue Y. Thermal
and morphological study of fractionated poly(3-hydroxybutyric
acid-co-3-hydroxypropionic acid). Macromolecular Chemistry and
Physics 1997;198:3539–57.
[317] Cao A, Kasuya K, Abe H, Doi Y, Inoue Y. Studies on comonomer
compositional distribution of the bacterial poly(3-hydroxybutyric
acid-co-3-hydroxypropionic acid)s and crystal and thermal char-
acteristics of their fractionated component copolyesters. Polymer
1998;39:4801–16.
[318] Cao A, Arai Y, Yoshie N, Kasuya KI, Doi Y, Inoue Y. Solid
structure and biodegradation of the compositionally fractionated
poly(3-hydroxybutyric acid-co-3-hydroxypropionic acid)s. Poly-
mer 1999;40:6821–30.
[319] Arai Y, Cao A, Yoshie N, Inoue Y. Studies on comonomer com-
positional distribution and its effect on some physical properties
of bacterial poly(3-hydroxybutyric acid-co-3-hydroxypropionic
acid). Polymer International 1999;48:1219–28.
[320] Wang Y, Ichikawa M, Cao A, Yoshie N, Inoue Y. Comonomer compo-
sition distribution of P(3HB-co-3HP)s produced by Alcaligenes latus
at several pH conditions. Macromolecular Chemistry and Physics
1999;200:1047–53.
[321] Yu F, Nakamura N, Inoue Y. Comonomer-unit compositional dis-
tribution and its effect on thermal and crystallization behavior of
bacterial poly(3-hydroxybutyrate-co-3-hydroxyhexanoate). Poly-
mer 2010;51:4408–18.
[322] Watanabe T, He Y, Fukuchi T, Inoue Y. Comonomer compositional
distribution and thermal characteristics of bacterially synthesized
poly(3-hydroxybutyrate-co-3-hydroxyhexanoate)s. Macromolec-
ular Bioscience 2001;1:75–83.
[323] Feng LD, Watanabe T, Wang Y, Kichise T, Fukuchi T, Chen G-Q, Doi
Y, Inoue Y. Studies on comonomer compositional distribution of
bacterial poly(3-hydroxybutyrate-co-3-hydroxyhexanoate)s and
thermal characteristics of their factions. Biomacromolecules
2002;3:1071–7.
[324] Asrar J, Valentin HE, Berger PA, Tran M, Padgette SR, Garbow
JR. Biosynthesis and properties of poly(3-hydroxybutyrate-
co-3-hydroxyhexanoate) polymers. Biomacromolecules
2002;3:1006–12.
[325] Doi Y, Segawa A, Kunioka M. Biosynthesis and characterization
of poly(3-hydroxybutyrate-co-4-hydroxybutyrate) in Alcaligenes
eutrophus. International Journal of Biological Macromolecules
1990;12:106–11.
[326] Shi FY, Ashby RD, Gross RA. Fractionation and characteriza-
tion of microbial polyesters containing 3-hydroxybutyrate and
4-hydroxybutyrate repeat units. Macromolecules 1997;30:2521–3.
[327] Ishida K, Wang Y, Inoue Y. Comonomer unit composition
and thermal properties of poly(3-hydroxybutyrate-co-4-
hydroxybutyrate)s biosynthesized by Ralstonia eutropha.
Biomacromolecules 2001;2:1285–93.
[328] Valentin HE, Berger PA, Gruys KJ, de Andrade Rodrigues MF,
Steinbüchel A, Tran M, Asrar J. Biosynthesis and character-
ization of poly(3-hydroxy-4-pentenoic acid). Macromolecules
1999;32:7389–95.
[329] Chen SS, Liu Q, Wang HH, Zhu B, Yu F, Chen G-Q, Inoue Y. Polymor-
phic crystallization of fractionated microbial medium-chain-length
polyhydroxyalkanoates. Polymer 2009;50:4378–88.
[330] Kato M, Bao HJ, Kang CK, Fukui T, Doi Y. Production of a novel
copolyester of 3-hydroxybutyric acid and medium chain length
3-hydroxyalkanaic acids by Pseudomonas sp 61-3 from sugars.
Applied Microbiology and Biotechnology 1996;45:363–70.
[331] Ashby RD, Solaiman DKY, Foglia TA. The synthesis of short-
and medium-chain-length poly(hydroxyalkanoate) mixtures from
glucose- or alkanoic acid-grown Pseudomonas oleovorans. Journal
of Industrial Microbiology and Biotechnology 2002;28:147–53.
[332] Yu F, Zhu B, Dong T, Inoue Y. Effect of comonomer-unit compo-
sitional distribution on thermal and crystallization behavior of
bacterial poly[(3-hydroxybutyrate)-co-(3-mercaptopropionate)].
Macromolecular Bioscience 2009;9:702–12.
[333] Kumagai Y, Kanesawa Y, Doi Y. Enzymatic degradation of micro-
bial poly(3-hydroxybutyrate) films. Die Makromolekulare Chemie
1992;193:53–7.
[334] Organ SJ. Phase separation in blends of poly(hydroxybutyrate) with
poly(hydroxybutyrate-co-hydroxyvalerate) – variation with blend
components. Polymer 1994;35:86–92.
[335] Saito M, Inoue Y, Yoshie N. Cocrystallization and phase
segregation of blends of poly(3-hydroxybutyrate) and
poly(3-hydroxybutyrate-co-3-hydroxyvalerate). Polymer
2001;42:5573–80.
[336] Yoshie N, Asaka A, Inoue Y. Cocrystallization and phase segregation
in crystalline/crystalline polymer blends of bacterial copolyesters.
Macromolecules 2004;37:3770–9.
[337] Yoshie N, Inoue Y. Cocrystallization and phase segregation in
blends of two bacterial polyesters. Macromolecular Symposium
2005;224:59–70.
[338] Saad GR. Blends of bacterial poly[(R)-3-hydroxybutyrate] with
oligo[(R,S)-3-hydroxybutyrate]-diol. Polymer International
2002;51:338–48.
[339] Pearce R, Jesudason J, Orts W, Marchessault RH. Blends of bacte-
rial and synthetic poly(␤-hydroxybutyrate) – effect of tacticity on
melting behavior. Polymer 1992;33:4647–9.
[340] Scandola M, Focarete ML, Adamus G, Sikorska W, Baranowska I,
Swierczek S, Gnatowski M, Kowalczuk M, Jedliński Z. Polymer
blends of natural poly(3-hydroxybutyrate-co-3-hydroxyvalerate)
and a synthetic atactic poly(3-hydroxybutyrate). Characteri-
zation and biodegradation studies. Macromolecules 1997;30:
2568–74.
[341] Hirota Y, Yoshie N, Ishii N, Kasuya KI, Inoue Y. Correlation between
solid-state structures and enzymatic degradability of cocrystallized
blends. Macromolecular Bioscience 2005;5:1094–100.
[342] Feng LD, Watanabe T, He Y, Wang Y, Kichise T, Fukuchi T. Phase
behavior and thermal properties for binary blends of bacterial
poly(3-hydroxybutyrate-co-3-hydroxyhexanoate)s with narrow-
comonomer-unit compositional distribution. Macromolecular Bio-
science 2003;3:310–9.
[343] de Koning D. Physical properties of bacterial poly((R)-
3-hydroxyalkanoates). Canadian Journal of Microbiology
1995;41(13):303–9.
[344] Noohom W, Jack KS, Martin D, Trau M. Understanding the roles of
nanoparticle dispersion and polymer crystallinity in controlling the
mechanical properties of HA/P(3HB)V nanocomposites. Biomedical
Materials 2009;4, 015003/1–13.
[345] Bloembergen S, Holden DA, Bluhm TL, Hamer GK, Marchessault
RH. Isodimorphism in synthetic poly(␤-hydroxybutyrate-co-␤-
hydroxyvalerate) – stereoregular copolyesters from racemic
␤-lactones. Macromolecules 1989;22:1663–9.
[346] Zhao W, Chen GQ. Production and characterization of
terpolyester poly(3-hydroxybutyrate-co-3-hydroxyvalerate-co-3-
hydroxyhexanoate) by recombinant Aeromonas hydrophila 4AK4
harboring genes phaAB. Process Biochemistry 2007;42:1342–7.
[347] Shimamura E, Scandola M, Doi Y. Microbial synthesis and char-
acterization of poly(3-hydroxybutyrate-co-3-hydroxypropionate).
Macromolecules 1994;160:4429–35.
[348] Lemos PC, Viana C, Salgueiro EN, Ramos AM, Crespo JPSG, Reis MAM.
Effect of carbon source on the formation of polyhydroxyalkanoates
(PHA) by a phosphate-accumulating mixed culture. Enzyme and
Microbial Technology 1998;22:662–71.
[349] Patel M, Gapes DJ, Newman RH, Dare PH. Physico-chemical
properties of polyhydroxyalkanoate produced by mixed-culture
nitrogen-fixing bacteria. Applied Microbiology and Biotechnology
2009;82:545–55.
[350] Carrasco F, Dionisi D, Martinelli A, Majone M. Thermal stabil-
ity of polyhydroxyalkanoates. Journal of Applied Polymer Science
2006;100:2111–21.
[351] Dobroth ZT, Hu SJ, McDonald AG, Coats ER. Polyhydroxybutyrate
synthesis on biodiesel wastewater using mixed microbial consortia.
Bioresource Technology 2011;102:3352–9.
[352] Iwata T, Tanaka T. Manufacturing of PHA as fibers. In: Chen G-Q,
editor. Plastics from bacteria: natural functions and applications.
Berlin; Heidelberg: Springer-Verlag; 2010. p. 257–82.
[353] Miller ND, Williams DF. On the biodegradation of
poly-␤-hydroxybutyrate (P(3HB)) homopolymer and poly-␤-
hydroxybutyrate-hydroxyvalerate copolymers. Biomaterials
1987;8:129–37.
 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
[354] Schmack G, Jehnichen D, Vogel R, Tändler B. Biodegradable fibers
of poly(3-hydroxybutyrate) produced by high-speed melt spin-
ning and spin drawing. Journal of Polymer Science Part B: Polymer
Physics 2000;38:2841–50.
[355] Iwata T. Strong fibers and films of microbial polyesters. Macro-
molecular Bioscience 2005;5:689–701.
[356] Iwata T, Aoyagi Y, Fujita M, Yamane H, Doi Y, Suzuki Y. Processing
of a strong biodegradable poly[(R)-3-hydroxybutyrate] fiber and
a new fiber structure revealed by micro-beam X-ray diffraction
with synchrotron radiation. Macromolecular Rapid Communica-
tions 2004;25:1100–4.
[357] Tanaka T, Yabe T, Teramachi S, Iwata T. Mechanical properties
and enzymatic degradation of poly[(R)-3-hydroxybutyrate] fibers
stretched after isothermal crystallization near Tg. Polymer Degra-
dation and Stability 2007;92:1016–24.
[358] Barham PJ, Keller A. The relationship between microstructure and
mode of fracture in polyhydroxybutyrate. Journal of Polymer Sci-
ence Part B: Polymer Physics 1986;24:69–77.
[359] Martinez-Salazar J, Sanchez-Cuesta M, Barham P, Keller A. Thermal
expansion and spherulite cracking in 3-hydroxybutyrate/3-
hydroxyvalerate copolymers. Journal of Materials Science Letters
1989;8:490–549.
[360] El-Hadi A, Schnabel R, Straube E, Muller G, Henning S. Cor-
relation between degree of crystallinity, morphology, glass
temperature, mechanical properties and biodegradation of poly(3-
hydroxyalkanoate) PHAs and their blends. Polymer Testing
2002;21:665–74.
[361] Kai WH, He Y, Inoue Y. Fast crystallization of
poly(3-hydroxybutyrate) and poly(3-hydroxybutyrate-co-3-
hydroxy-valerate) with talc and boron nitride as nucleating
agents. Polymer International 2005;54:780–9.
[362] Kai WH, He Y, Asakawa N, Inoue Y. Effect of lignin particles as
a nucleating agent on crystallization of poly(3-hydroxybutyrate).
Journal of Applied Polymer Science 2004;94:2466–74.
[363] He Y, Inoue Y. ␣-Cyclodextrin-enhanced crystallization of poly(3-
hydroxybutyrate). Biomacromolecules 2003;4:1865–7.
[364] Jacquel N, Tajima K, Nakamura N, Miyagawa T, Pan PJ, Inoue
Y. Effect of orotic acid as a nucleating agent on the crystalliza-
tion of bacterial poly(3-hydroxybutyrate-co-3-hydroxyhexanoate)
copolymers. Journal of Applied Polymer Science 2009;114:
1287–94.
[365] Volova T. Polyhydroxyalkanoates – plastic materials of the 21st cen-
tury: production, properties, applications. Hauppauge, NY: Nova
Publishers; 2004, 282 pp.
[366] Choi JS, Park WH. Effect of biodegradable plasticizers on thermal
and mechanical properties of poly(3-hydroxybutyrate). Polymer
Testing 2004;23:455–60.
[367] Bordes P, Pollet E, Averous L. Nano-biocomposites: biodegrad-
able polyester/nanoclay systems. Progress in Polymer Science
2009;34:125–55.
[368] Galego N, Rozsa C, Sanchez R, Fung J, Vazquez A, Tomas JS. Charac-
terization and application of poly(␤-hydroxyalkanoates) family as
composite biomaterials. Polymer Testing 2000;19:485–92.
[369] Sherman LM. Enhancing biopolymers: additives are needed
for toughness, heat resistance and processability; 2012,
http://www.allbusiness.com/chemicals/plastics-rubber-industry-
resins/11421097-1.html#ixzz1lKy5XtiG [accessed 31 January
2012].
[370] Barham PJ, Organ SJ. Mechanical properties of
polyhydroxybutyrate–hydroxybutyrate–hydroxyvalerate
copolymer blends. Journal of Materials Science 1994;29:1676–9.
[371] Satoh H, Yoshie N, Inoue Y. Hydrolytic degradation of blends
of poly(3-hydroxybutyrate) with poly(3-hydroxybutyrate-co-3-
hydroxyvalerate). Polymer 1994;35:286–90.
[372] Yang F, Li Z, Qiu Z. Miscibility and crystallization
behavior of biodegradable poly(3-hydroxybutyrate-co-3-
hydroxyvalerate)/phenolic blends. Journal of Applied Polymer
Science 2012;123:2781–6.
[373] Ha CS, Cho WJ. Miscibility, properties, and biodegradability of
microbial polyester containing blends. Progress in Polymer Science
2002;27:759–809.
[374] Yu L, Dean K, Li L. Polymer blends and composites from renewable
resources. Progress in Polymer Science 2006;31:576–602.
[375] Zhang JW, McCarthy S, Whitehouse R. Reverse temperature injec-
tion molding of BiopolTM and effect on its properties. Journal of
Applied Polymer Science 2004;94:483–91.
[376] Chua H, Hu PHF, Ma CK. Accumulation of biopolymers in acti-
vated sludge biomass. Applied Biochemistry and Biotechnology
1999;77(9):389–99.
441
[377] Punrattanasin W. The utilization of activated sludge polyhydrox-
yalkanoates for the production of biodegradable plastics. PhD
Dissertation. Environmental Science and Engineering, Blacksburg,
Virginia, USA: Virginia Polytechnic Institute and State University;
2001. 120 pp.
[378] Serafim LS, Lemos PC, Torres C, Reis MAM, Ramos AM. The influence
of process parameters on the characteristics of polyhydroxyalka-
noates produced by mixed cultures. Macromolecular Bioscience
2008;8:355–66.
[379] Morgan-Sagastume F, Karlsson A, Johansson P, Pratt S, Boon N,
Lant P, Werker A. Production of polyhydroxyalkanoates in open,
mixed cultures from a waste sludge stream containing high lev-
els of soluble organics, nitrogen and phosphorus. Water Research
2010;44:5196–211.
[380] Chanprateep S, Buasri K, Muangwong A, Utiswannakul P.
Biosynthesis and biocompatibility of biodegradable poly(3-
hydroxybutyrate-co-4-hydroxybutyrate). Polymer Degradation
and Stability 2010;95:2003–12.
[381] de Koning GJM, Lemstra PJ. Crystallization phenomena in bacterial
poly[(R)-3-hydroxybutyrate]. 2. Embrittlement and rejuvenation.
Polymer 1993;34:4089–94.
[382] Doi Y, Kitamura S, Abe H. Microbial synthesis and char-
acterization of poly(3-hydroxybutyrate-co-3-hydroxyhexanoate).
Macromolecules 1995;28:4822–8.
[383] Gogotov IN, Gerasin VA, Knyazev YV, Antipov EM, Barazov SK. Com-
posite biodegradable materials based on polyhydroxyalkanoate.
Applied Biochemistry and Microbiology 2010;46:607–13.
[384] Gregorova A, Wimmer R, Hrabalova M, Koller M, Ters T, Mundigler
N. Effect of surface modification of beech wood flour on mechani-
cal and thermal properties of poly(3-hydroxybutyrate)/wood flour
composites. Holzforschung 2009;63:565–70.
[385] Howells ER. Opportunities in biotechnology for the chemical
industry. Journal of the Society of Chemical Industry, London
1982;8:508–11.
[386] King PP. Biotechnology – an industrial view. Journal of Chemical
Technology and Biotechnology 1982;32:2–8.
[387] Mitomo H, Barham PJ, Keller A. Temperature-dependence
of mechanical-properties of poly(␤-hydroxybutyrate-␤-
hydroxyvalerate). Polymer Communications 1988;29:112–5.
[388] Savenkova L, Gercberga Z, Bibers I, Kalnin M. Effect of 3-hydroxy
valerate content on some physical and mechanical properties of
polyhydroxyalkanoates produced by Azotobacter chroococcum. Pro-
cess Biochemistry 2000;36:445–50.
[389] Cai ZJ, Yang G. Optical nanocomposites prepared by incorporating
bacterial cellulose nanofibrils into poly(3-hydroxybutyrate). Mate-
rials Letters 2011;65:182–4.
[390] Wang SA, Ma PM, Wang RY, Wang SF, Zhang Y,
Zhang YX. Mechanical, thermal and degradation
properties of poly(d,l-lactide)/poly(hydroxybutyrate-co-
hydroxyvalerate)/poly(ethylene glycol) blend. Polymer
Degradation and Stability 2008;93:1364–9.
[391] Wang W, Zhang Y, Zhu MF, Chen YM. Effect of graft modification
with poly(N-vinylpyrrolidone) on thermal and mechanical proper-
ties of poly(3-hydroxybutyrate-co-3-hydroxyvalerate). Journal of
Applied Polymer Science 2008;109:1699–707.
[392] Modi SJ. Assessing the feasibility of poly(3-hydroxybutyrate-co-3-
hydroxyvalerate) (P(3HB)V) and poly(lactic acid) for potential food
packaging applications. MS Dissertation. Ohio State University.
http://rave.ohiolink.edu/etdc/view?acc num=osu1268921056;
2010. 123 pp.
[393] Avella M, La Rota G, Martuscelli E, Raimo M, Sadocco P, Ele-
gir G, Riva R. Poly(3-hydroxybutyrate-co-3-hydroxyvalerate) and
wheat straw fibre composites: thermal, mechanical proper-
ties and biodegradation behaviour. Journal of Materials Science
2000;35:829–36.
[394] Fei B, Chen C, Chen S, Peng S, Zhuang Y, An Y, Dong L.
Crosslinking of poly[(3-hydroxybutyrate)-co-(3-hydroxyvalerate)]
using dicumyl peroxide as initiator. Polymer International 2004;53:
937–43.
[395] Zhang HF, Ma L, Wang ZH, Chen GQ. Biosynthesis and charac-
terization of 3-hydroxyalkanoate terpolyesters with adjustable
properties by Aeromonas hydrophila. Biotechnology and Bioengi-
neering 2009;104:582–9.
[396] Chen GX, Hao GJ, Guo TY, Song MD, Zhang BH. Crystallization
kinetics of poly(3-hydroxybutyrate-co-3-hydroxyvalerate)/clay
nanocomposites. Journal of Applied Polymer Science
2004;93:655–61.
[397] Gursel I, Balcik C, Arica Y, Akkus O, Akkas N, Hasirci V. Synthe-
sis and mechanical properties of interpenetrating networks of
442 B. Laycock et al. / Progress in Polymer Science 39 (2014) 397–442
polyhydroxybutyrate-co-hydroxyvalerate and polyhydroxyethyl
methacrylate. Biomaterials 1998;19:1137–43.
[398] Jiang L, Huang JJ, Qian J, Chen F, Zhang JW, Wolcott MP,
Zhu Y. Study of poly(3-hydroxybutyrate-co-3-hydroxyvalerate)
(P(3HB)V)/bamboo pulp fiber composites: effects of nucleation
agent and compatibilizer. Journal of Polymers and the Environment
2008;16:83–93.
[399] Avella M, Bogoeva-Gaceva G, Buzarovska A, Errico ME, Gentile
G, Grozdanov A. Poly(3-hydroxybutyrate-co-3-hydroxyvalerate)-
based biocomposites reinforced with kenaf fibers. Journal of
Applied Polymer Science 2007;104:3192–200.
[400] Bhardwaj R, Mohanty AK, Drzal LT, Pourboghrat F, Misra M. Renew-
able resource-based green composites from recycled cellulose
fiber and poly(3-hydroxybutyrate-co-3-hydroxyvalerate) bioplas-
tic. Biomacromolecules 2006;7:2044–51.
[401] Dagnon KL, Chen HH, Lnnocentini-Mei LH, D’Souza NA.
Poly[(3-hydroxybutyrate)-co-(3-hydroxyvalerate)]/layered
double hydroxide nanocomposites. Polymer International
2009;58:133–41.
[402] Saito Y, Doi Y. Microbial synthesis and properties of poly(3-
hydroxybutyrate-co-4-hydroxybutyrate) in Comamonas
acidovorans. International Journal of Biological Macromolecules
1994;16:99–104.
[403] Nakamura K, Kamiya N, Sakurai M, Inoue Y, Chujo R. A
molecular mechanics study on conformations of bacterial
polyester poly-(3-hydroxybutyrate-co-3-hydroxyvalerate). Poly-
mer 1992;33:817–22.
[404] Tanadchangsaeng N, Kitagawa A, Yamamoto T, Abe H, Tsuge
T. Identification, biosynthesis, and characterization of poly-
hydroxyalkanoate copolymer consisting of 3-hydroxybutyrate
and 3-hydroxy-4-methylvalerate. Biomacromolecules 2009;10:
2866–74.
[405] Hassan MK, Abou-Hussein R, Zhang XJ, Mark JE, Noda I. Biodegrad-
able copolymers of 3-hydroxybutyrate-co-3-hydroxyhexanoate
(NodaxTM ), including recent improvements in their mechan-
ical properties. Molecular Crystals and Liquid Crystals
2006;447:341–62.
[406] Shen L, Haufe J, Patel MK. Product overview and market projection
of emerging biobased plastics (PROBIP 2009). Commissioned by
European Polysaccharide Network of Excellence (EPNOE) and Euro-
pean Bioplastics. Report No.: NWS-E-2009-32. Science, Technology
and Society Group, Copernicus Institute for Sustainable Develop-
ment and Innovation Utrecht The Netherlands: Utrecht University;
2009.
[407] Zoller DL, Johnston MV. Microstructures of butadiene copolymers
determined by ozonolysis/MALDI mass spectrometry. Macro-
molecules 2000;33:1664–70.
[408] Semler JJ, Jhon YK, Tonelli A, Beevers M, Krishnamoorti R, Genzer
J. Facile method of controlling monomer sequence distributions in
random copolymers. Advanced Materials 2007;19:2877–83.