Thermoplastic polyester amides derived from oleic acid

Article abstract of DOI:10.1016/j.polymer.2011.08.002

Three lipid-based Polyester Amides (PEAs) with varying ratios of ester and amide linkages were synthesized. Oleic acid was used as the starting material to produce the intermediates, characterized by MS and NMR, used for polymerization. PEAs were characterized by FTIR and GPC. The PEAs were constrained to have similar number average molecular weights, in the 2 × 10⁴ range, thereby enabling comparison of their physical properties from a structural perspective. The thermal behavior of the polymers was assessed by DSC, DMA and TGA. Thermal degradation was not affected by ester/amide ratios, but T_g increased non-linearly with decreasing ester/amide ratios and correlated with hydrogen-bond density and repeating unit chain length. Crystallinity was studied by XRD and DSC. Degree of crystallization and multiple melting behavior as a function of cooling kinetics were explained well by hydrogen-bond density, repeating unit chain length and density of ester moieties. Mechanical properties were investigated by DMA and Tensile Analysis, with a non-linear increase of storage and tensile moduli recorded as a function of decreasing ester/amide ratios. The findings suggest how approaches to the synthesis of lipid-based PEAs can be targeted to the delivery of specific physical properties.

Full text of DOI:10.1016/j.polymer.2011.08.002

Polymer 52 (2011) 4503e4516
Contents lists available at ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Thermoplastic polyester amides derived from oleic acid
a
a
a
a,b,
*
Jiaqing Zuo , Shaojun Li , Laziz Bouzidi , Suresh S. Narine
a
Trent Biomaterials Research Program, Department of Physics & Astronomy, Trent University, 1600 West Bank Drive, Peterborough, Ontario K9J 7B8, Canada
Trent Biomaterials Research Program, Department of Chemistry, Trent University, 1600 West Bank Drive, Peterborough, Ontario K9J 7B8, Canada
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 29 June 2011
Received in revised form
Three lipid-based Polyester Amides (PEAs) with varying ratios of ester and amide linkages were
synthesized. Oleic acid was used as the starting material to produce the intermediates, characterized by
MS and NMR, used for polymerization. PEAs were characterized by FTIR and GPC. The PEAs were con-
29 July 2011
4
strained to have similar number average molecular weights, in the 2 ꢀ 10 range, thereby enabling
Accepted 4 August 2011
Available online 8 August 2011
comparison of their physical properties from a structural perspective. The thermal behavior of the
polymers was assessed by DSC, DMA and TGA. Thermal degradation was not affected by ester/amide
g
ratios, but T increased non-linearly with decreasing ester/amide ratios and correlated with hydrogen-
Keywords:
bond density and repeating unit chain length. Crystallinity was studied by XRD and DSC. Degree of
crystallization and multiple melting behavior as a function of cooling kinetics were explained well by
hydrogen-bond density, repeating unit chain length and density of ester moieties. Mechanical properties
were investigated by DMA and Tensile Analysis, with a non-linear increase of storage and tensile moduli
recorded as a function of decreasing ester/amide ratios. The findings suggest how approaches to the
synthesis of lipid-based PEAs can be targeted to the delivery of specific physical properties.
Ó 2011 Elsevier Ltd. All rights reserved.
Lipid-based polyester amide
Ester to amide ratio
Thermal and mechanical properties
1. Introduction
surgical materials, tissue engineering materials as well as carriers
in drug delivery systems [4,6]. Most of today’s plastics and poly-
Polyester amides (PEAs) have been extensively studied for
several years; because they generally demonstrate desired prop-
erties of both polyesters and polyamides [1]. PEAs are mainly
developed from three building blocks: amino acids, diols and diacyl
chlorides [2e5]. Polyamides generally demonstrate preferred
mechanical and processing properties, such as thermal stability,
tensile strength and biocompatibility, but are generally not biode-
gradable [6]. Polyesters demonstrate varying degrees of biode-
gradability; however, lack of the preferred physical properties
demonstrated for example by polyamides limits their applications.
Therefore when these two different types of linkages are combined
in the main chain of one single polymer, the biodegradability
caused by the ester linkage and the enhanced thermal and
mechanical properties caused by amide-induced intermolecular
hydrogen bonds can be combined in one polymer [7], giving rise to
a superior material which could be applied in a wide variety of
areas.
meric materials are made from petroleum-based products.
However, in the last few years, environmental issues and concerns
have drawn significant attention to petroleum-based products,
especially in polymeric materials. Petroleum-based polymers are
generally non-degradable, and are implicated in the pollution of
the natural environment. Petroleum is also regarded as a finite and
dwindling resource. The desire to obtain various products from
non-toxic natural resources, and use bio-based renewable products
as alternatives for the petroleum-based products is increasing in
our society [8e17].
Several investigations have been conducted on producing PEAs
from bio-based sustainable resources, such as linseed oil, nahar
seed oil and pongamia glabra oil [18e20]. However, the building
blocks chosen to connect the lipid-based monomers were not from
natural resources, and were mostly toxic. The toxicity of these PEAs
limits their applications and motivates the development of alter-
native building blocks. Several other fatty acid-based PEAs were
studied and synthesized from sustainable resources; however, the
fatty acid chosen, e.g., gallic acid [21], was much less abundant than
oleic acid.
Biomedical applications is one of the most important areas to
which PEAs have generally been applied; this includes absorbable
This present study is targeted at the development of lipid-based
polymers that would be tough, thermally stable and biodegradable,
yet be safe to use in a wide variety of areas. In particular, intro-
ducing amide units to enhance the material’s cohesion thorough
*
Corresponding author. Trent Biomaterials Research Program, Department of
Physics & Astronomy, Trent University, 1600 West Bank Drive, Peterborough,
Ontario, Canada, K9J 7B8. Tel.: þ1 705 748 1011; fax: þ1 705 748 1652.
E-mail address: sureshnarine@trentu.ca (S.S. Narine).
0
032-3861/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymer.2011.08.002

4504
J. Zuo et al. / Polymer 52 (2011) 4503e4516
hydrogen bonding is proposed. In order to understand the struc-
tureeproperty relationship and assess any underlying trend, three
PEAs with different ratios of ester to amide groups, i.e.,1:1 (PEA (I)),
0.03 mol) was added, followed by adding DCC (49.4 g,
0.24 mol) slowly to the reaction mixture. The ice bath was
removed and the reaction was stirred at room temperature
for 24 h. The resulting mixture was filtered to remove the
solid. The crude products were collected by evaporating
DCM under vacuum. Column chromatography was used to
purify the product (hexane/ethyl acetate 30:1) (Yield: 88%).
Step 2. Di-ester 3 (30.2 g, 0.05 mol) was dissolved in 300 mL of
anhydrous ethanol in a three-necked round bottom flask
1
:2 (PEA (II)) and 1:3 (PEA (III)) were prepared. The lipid-based,
non-toxic components, oleic acid-derived diol, amino acid and
azelaoyl chloride (also derived from oleic acid) were chosen as the
building blocks of the new polymeric materials. Furthermore, the
synthesis procedures were simple and performed under mild
conditions. The PEA samples were synthesized with similar
4
4
ꢁ
number average molecular weights, in the 2.08 ꢀ 10 to 2.26 ꢀ 10
and cooled to ꢂ20 C using an ice salt bath. Ozone was
range, to enable comparison of their physical properties. Wide
angle X-ray diffraction (WAXD), DSC, TGA, DMA and tensile tech-
niques were used to investigate the crystal structure, and thermal
and mechanical properties of the polymers.
bubbled into the solution with a flow rate of 5 L/min. The
reaction was monitored by thin layer chromatography
(TLC) until the starting material was gone. After the reac-
tion, nitrogen was purged through the mixture for 20 min
to remove the extra ozone in the flask. Next, 3.8 g NaBH
0.1 mol) was slowly added into the ozonolysis mixture.
The reaction was stopped after 4 h, and then water and
dilute hydrochloride acid were added into the reaction
4
(
2
. Experimental
2.1. Materials
4
mixture to eliminate the extra NaBH . The resulting
mixture was extracted by 2 ꢀ 200 mL of ethyl acetate. The
All reagents, Oleic Acid (90% purity), Potassium hydroxide
ethyl acetate phase was washed by brine, and dried over
(
KOH), Potassium permanganate (KMnO
4
), Lithium aluminum
2 4
Na SO . The crude products were collected by removing
hydride (LiAlH
4
), 1,3-propane diol, 4-(Dimethylamino) pyridine
0
the solvent under vacuum. The desired product was puri-
fied by recrystallization using ethyl acetate and hexanes
with a ratio around 1:3 (Yield: 72%).
(
DMAP), N, N -Dicyclohexylcarbodiimide (DCC), sodium borohy-
4
dride (NaBH ), Boc-Ala-OH (N-tert-butoxycarbonyl protected
alanine), Trifluoroacetic acid (TFA), Azelaoyl chloride, N-methyl-2-
pyrrolidone (NMP), and propylene oxide, were purchased from
SigmaeAldrich.
1H NMR (CDCl
.64e3.61 (t, 4H, CH
m, 2H, CH CH OCOR), 1.63e1.52 (m, 8H, CH
40 6
OH), 1.36e1.27 (m, 16H, CH ). MS (ESI): calcd for C21H O
, 400 MHz)
d
(ppm): 4.16e4.12 (t, 4H, CH
COOR), 1.99e1.93
CH COOR and
3
2
OCOR),
3
(
CH
3
2
OH), 2.31e2.27 (t, 4H, CH
2
2
2
2
2
2
2
.2. Synthesis of diols from oleic acid
.2.1. 1,9-Nonanediol
2
CH
2
2
þ
88.28, found m/z 411.3 ([M þ Na] ).
1
,9-Nonanediol was synthesized by a two-step procedure,
2
.2.3. Tetra-ester diol
Tetra-ester diol was synthesized by a two-step procedure as
starting from oleic acid (Scheme 1). First KMnO was used to
4
oxidize the double bonds of the oleic acid, forming azelaic acid,
following procedures in the literature [22] (Yield: 80%). 1,9-
Nonanediol was then prepared according to the literature by
shown in Scheme 3. First, tetra-ester 5 was synthesized by reacting
the di-ester diol and oleic acid (Yield: 86%); the esterification
reaction between diol and oleic acid has been described in the
synthesis procedure of di-ester 3. Then tetra-ester diol was
synthesized by breaking the double bonds and changing them to
reducing the azelaic acid with LiAlH
4
[23] (Yield: 88%).
1
H NMR (CDCl
.59e1.55 (m, 4H, CH
for C
3
, 400 MHz)
d
(ppm): 3.67e3.62 (t, 4H, CH
2
OH),
). MS (ESI): calcd
1
2
CH
2
OH), 1.32 (m, 10H, CH
2
hydroxide groups, using ozonolysis and NaBH
been described in the synthesis of di-ester diol 4 (Yield: 50%).
4
. This reaction has
þ
9
20
H O
2
160.15, found m/z 183.0 ([M þ Na] ).
1
H NMR (CDCl
OCOR), 3.66e3.61 (t, 4H, CH
.99e1.94 (m, 2H, ROCOCH CH
CH CH COOR and CH CH OH), 1.32e1.28 (m, 24H, CH
calcd for C39
10 700.51, found m/z 723.8 ([M þ Na] ).
3
, 400 MHz)
OH), 2.32e2.26 (t, 8H, CH
CH OCOR), 1.61e1.53 (m, 24H,
). MS (ESI):
d
(ppm): 4.17e4.03 (m, 8H,
2
.2.2. Di-ester diol
Di-ester diol was synthesized by a two-step procedure as shown
in Scheme 2.
CH
1
2
2
2
COOR),
2
2
2
2
2
2
2
2
þ
72
H O
Step 1. 1, 3-propane diol (7.6 g, 0.1 mol) and oleic acid (69 g, 90%
purity, 0.22 mol) were dissolved in 200 mL of dichloro-
methane (DCM) in a round bottom flask. The reaction
2
2
.3. Polymerization
ꢁ
mixture was kept at 0 C using an ice bath. DMAP (3.6 g,
.3.1. Syntheis of diamines
Because three polymers were synthesized in this study,
Oleic Acid
a general synthesis procedure to prepare polymers is shown in
Scheme 4 and described as follows. First, a selected diol (10 mmol)
and Boc-Ala-OH (22 mmol) were dissolved in DCM in a round
COOH
KMnO4
ꢁ
bottom flask. The flask was kept at 0 C in an ice bath. DMAP
(
3 mmol) was then added to the mixture. DCC (24 mmol) was
HOOC
HO
COOH
slowly added over 30 min. The ice bath was removed and the
reaction was stirred at room temperature for 24 h, and the resulting
mixture was then filtered. The crude products were collected by
removing DCM under vacuum. Column chromatography was used
to purify the product.
TFA was added drop wise to this product in DCM in a round
bottom flask. The reaction was stirred at room temperature until
the starting materials had all reacted; the reaction was monitored
by TLC. After the reaction, DCM was evaporated by rotary
Azelaic acid
1
LiAlH4
OH
2
1
, 9- nonanediol
Scheme 1. 1, 9-Nonanediol from oleic acid.

J. Zuo et al. / Polymer 52 (2011) 4503e4516
4505
Oleic Acid
COOH
HO
O
OH
O
O
O
Di-ester
3
O3
NaBH4
O
O
HO
O
O
OH
Di-ester diol
4
Scheme 2. Di-ester diol synthesized from oleic acid.
evaporation, and the residue was neutralized by a saturated sodium
bicarbonate (NaHCO ) solution to a pH value of 8 or 9. The diamine
was extracted by chloroform (CHCl ) and dried over Na SO . The
solvent was evaporated by reduced pressure and the residue was
purified by column chromatography with ethyl acetate, methanol
and ammonium hydroxide (100/10/1).
2.3.2. Polymerization
3
PEA was synthesized by conventional low-temperature solution
polymerization, and propylene oxide was used as a hydrogen
chloride acceptor in this reaction [24]. The polymerization proce-
dure is described here only for PEA (I). Diamine (I) (4.00 g,
7.55 mmol) was added in a round bottomed flask with 24 mL of
NMP under nitrogen gas; the reaction mixture was cooled at 0 C
using an ice bath. Propylene oxide (2.8 mL) and 1.70 g (7.55 mmol)
dichloride (Azelaoyl chloride) was added to the solution succes-
3
2
4
ꢁ
Diamine (I) was synthesized from 1,9-Nonanediol (Yield: 85%).
1
H NMR (CDCl
.63e3.53 (m, 2H, NH2CHCOOR), 2.13e2.06 (m, 4H, CH
m, 4H, CH ), 1.38e1.23 (m, 12H, CH , CH ).
3
, 400 MHz)
d
(ppm): 4.13e4.06 (m, 4H, CH
2
OCOR),
), 1.65e1.58
3
(
2
ꢁ
2
2
3
sively. The solution was stirred at 0 C for half an hour, and then the
ice bath was removed. After reacting at room temperature for 28 h,
the reaction mixture was precipitated into 200 mL of aqueous
1
Diamine (II) was synthesized from di-ester diol (Yield: 89%). H
NMR (CDCl
.53e3.47 (m, 2H, NH2CHCOOR), 2.28e2.24 (t, 4H, CH
.97e1.89 (m, 2H, ROCOCH CH CH
).
Diamine (III) was synthesized from tetra-ester diol (Yield: 93%).
H NMR (CDCl , 400 MHz): 4.15e4.01 (m, 12H, CH OCOR), 3.61 (m,
H, NHCHCOOR), 2.29e2.25 (m, 8H, CH COOR), 1.96e1.93 (m, 2H,
3
, 400 MHz)
d
(ppm): 4.12e4.03 (m, 8H, CH
2
OCOR),
3
2
COOR),
3
methanol (CH OH:H
2
O ¼ 3:1 v/v). The polymer was settled for 12 h.
1
2
2
2
OCOR), 1.61e1.56 (m, 12H, CH
2
),
PEA was then collected by vacuum filtration, washed with water
and dried in a vacuum oven at 50 C for 2 days.
ꢁ
1
.32e1.28 (m, 18H, CH2, CH
3
1
ꢂ1
3
2
2.3.2.1. PEA (I). IR (cm ): 3306 (NeH stretching), 2926, 2852 (CeH
stretching), 1736 (ester carbonyl stretching), 1643 (amide carbonyl
stretching), 1539 (CeNeH deformation). H NMR (CDCl
2
2
1
ROCOCH
0H, CH
2
CH
, CH
2
CH
).
2
OCOR), 1.60e1.54 (m, 24H, CH
2
), 1.37e1.22 (m,
3
, 400 MHz)
3
2
3
d
(ppm): 6.37 (d, NH), 4.59e4.52 (m, NHCHCOOR), 4.15e4.02 (m,
O
O
O
HO
O
OH
O
OH
O
O
O
O
O
O
O
O
Tetra-ester
5
O
3
NaBH4
O
O
O
O
HO
O
O
O
O
OH
Tetra-ester diol
6
Scheme 3. Tetra-ester diol synthesized from di-ester diol.

4506
J. Zuo et al. / Polymer 52 (2011) 4503e4516
HO
O
OH
O
O
H
N
COOH
HO
O
O
O
OH
Boc
H
N
COOH
Boc
TFA
O
TFA
H
2
N
NH
2
O
O
O
O
O
O
H2N
NH2
O
O
O
O
O
Cl
O
O
Cl
Cl
Cl
O
O
H
H
N
N
H
N
O
O
O
O
H
N
O
O
n
O
O
O
O
O
O
O
O n
PEA (I)
PEA (II)
O
O
O
O
O
O
HO
O
O
O
O
O
O
OH
H
N
COOH
Boc
TFA
O
O
O
O
H
2
N
NH
2
O
O
O
O
O
O
O
Cl
Cl
O
O
O
O
H
N
O
O
H
N
O
O
O
O
O
n
O
O
PEA (III)
Scheme 4. Polymerization process of PEAs.
CH
2
OCOR), 2.21e2.18 (m, CH
2
COOR), 1.61e1.60 (m, CH
). C NMR (CDCl , 300 MHz) (ppm): 173.5 (OeC]
O), 173.0 (NHeC]O), 65.8 (COOCH ), 48.2 (NHCHCOOR), 36.5
CH C]O), 29.5, 29.4, 29.2, 29.1, 28.7, 25.9, 25.7 (CH ), 18.8 (CH ).
Yield: 81%.
2
), 1.38e1.29
400 NMR spectrometer (Varian, Inc., Walnut Creek, U.S.A.). Mass
spectra were acquired on a Quattro LC (Micromass) electrospray
ionization (ESI) mass spectrometer with a syringe pump (Harvard).
FTIR spectra were measured with a Thermo Scientific Nicolet 380
FTIR spectrometer (Thermo Electron Scientific Instruments LLC,
U.S.A.) fitted with a PIKE MIRacleÔ attenuated total reflectance
(ATR) system (PIKE Technologies, Madison, WI, U.S.A.). The samples
for FTIR testing were made into 0.5 mm thick films and placed onto
the ATR crystal area and held in place by the pressure arm. The
molecular weight and distribution were determined by Gel
Permeation Chromatography (GPC) (Waters, MA). The test was
carried out with a Waters e2695 pump, Waters 2414 refractive
13
(
m, CH
2
and CH
3
3
d
2
(
2
2
3
ꢂ1
1
2
.3.2.2. PEA (II). IR (cm ): 3311, 2928, 2852, 1738, 1647, 1537. H
NMR (CDCl , 400 MHz) (ppm): 6.15e6.13 (d, NH), 4.55e4.48 (m,
NHCHCOOR), 4.11e4.01 (m, CH OCOR), 2.28e2.12 (t, CH COOR),
),1.56e1.48 (m, CH ), 1.34e1.24 (m, CH and CH ).
(ppm): 173.9, 173.5 (OeC]O), 172.9
), 48.2 (NHCHCOOR), 36.6, 34.4
3
d
2
2
1
.93e1.87 (m, CH
C NMR (CDCl
2
2
2
3
1
3
3
, 300 MHz) d
(
(
(
NHeC]O), 65.8, 61.0 (COOCH
2
CH
CH
2
3
C]O), 29.3, 29.2, 29.1, 28.7, 28.2, 25.9, 25.6, 25.1 (CH
). Yield: 78%.
2
), 18.9
index detector and a Styragel HR5E column (5 mm). Chloroform was
used as eluent with a flow rate of 1 mL/min. The sample was made
with a concentration of 4 mg/mL, and the injection volume was
30 ml for each sample. Polystyrene (PS) Standards were used to
ꢂ1
1
2
.3.2.3. PEA (III). IR (cm ): 3311, 2928, 2854, 1732, 1645, 1537. H
NMR (CDCl
NHCHCOOR), 4.14e4.01 (m, CH
3
, 400 MHz)
d
(ppm): 6.13e6.12 (d, NH), 4.58e4.54 (m,
OCOR), 2.29e2.16 (t, CH COOR),
), 1.60e1.55 (m, CH ),1.38e1.28 (m, CH and CH ).
(ppm): 174.1, 173.9, 173.6 (OeC]O),
), 48.2 (NHCHCOOR),
C]O), 29.4, 29.3, 29.2, 29.1, 28.8, 28.7, 28.2,
6.1, 26.0, 25.6, 25.1 (CH ), 18.9 (CH ). Yield: 76%.
calibrate the curve.
2
2
Thermogravimetric Analysis (TGA) was carried out using a Q500
(TA instrument, New Castle, DE, U.S.A.) following the ASTM D3850-
94 standard to test the thermal stability of the synthesized PEAs.
The polymer samples were loaded in the form of powders. Samples
were heated from room temperature to 700 C under dry nitrogen
at a constant heating rate of 10 C/min.
1
.97e1.91 (m, CH
C NMR (CDCl
2
2
2
3
1
3
3
, 300 MHz) d
1
3
2
72.9 (NHeC]O), 65.8, 64.6, 61.0 (COOCH
6.6, 34.5, 34.4 (CH
2
ꢁ
2
ꢁ
2
3
The DSC measurements have been carried out on a temperature
modulated DSC Q200 model (TA Instruments, New Castle, DE)
equipped with a refrigerated cooling system. During the heating
process, measurements were performed with a modulation
2
.4. Characterization techniques
1H NMR was used to qualitatively analyze the products, and was
ꢁ
recorded at a larmor frequency of 400 MHz, using a Varian Unity
amplitude of 1 C/min and a modulation period of 60 s. The

J. Zuo et al. / Polymer 52 (2011) 4503e4516
4507
measurements were performed at least in triplicate following the
ASTM E1356-03 standard procedure. Approximately 5.0e10.0
respectively. The ratio of the 6.37e6.12 to 4.15e4.01 ppm peak
integrals was 1:2 for PEA (I), 1:4 for PEA (II) and 1:6 for PEA (III).
This is consistent with the actual ratio of ester to amide groups in
the PEA structures since the integration of the 6.37e6.12 ppm peak
provides the amount of the amide linkages and the integration of
the 4.15e4.01 ppm provides twice the amount of ester linkages.
(
ꢃ0.1) mg of sample was placed and sealed in an aluminum DSC
pan. An empty aluminum pan was used as a reference and the
experiments were performed under a nitrogen flow of 50 mL/min.
The “TA Universal Analysis” software coupled with a method
developed by our group [25] was used to analyze the data and
13
The C NMR results were also fully consistent with the actual
ester to amide ratios in the PEA structures as revealed by the type of
extract the main characteristics, i.e., T
transition, onset, offset, and peak maximum temperature, respec-
tively) and enthalpy, H. The characteristics of the non-resolved
g m
, TOn, TOff and T (glass
13
ester and amide observed. The analysis of the C NMR peaks which
represent the carbon atom in ester carbonyl (in the range of
174.1e173.5 ppm), and the peak which represent the carbon atom
in the amide carbonyl (at 172.9 ppm or 173.0 ppm) indicated that
only one type of ester and one type of amide are shown in the
spectrum of PEA (I), two types of ester and one type of amide in the
spectrum of PEA (II) and three types of ester and one type of amide
in PEA (III).
D
individual peaks and shoulder signals were estimated using the
first and second derivatives of the signal and a simple decompo-
sition of the signal into its obvious main components.
A first DSC heating cycle was performed on the preformed PEA.
In this experiment, the sample which was ambient cooled was
ꢁ
ꢁ
heated from room temperature (10 C/min) to 140 C to record the
development of the phases preformed. In a second set of experi-
The IR spectra fully confirmed the anticipated chemical struc-
tures. As can be seen in Fig. 2 representing the IR of the three PEAs
ꢁ
ments, the sample was equilibrated at 140 C, and held at that
ꢂ1
temperature for 5 min to erase the thermal history; then cooled
in the 4000e700 cm range, the absorption bands characteristic of
ꢁ
ꢁ
ꢂ1
down to ꢂ60 with a cooling rate of 5 C/min. In the heating cycle,
the ester carbonyl stretching vibration are observed at 1736 cm
,
ꢁ
the sample was heated with a constant heating rate of 3 C/min
and the infrared bands characteristic of amide groups, i.e., NeH
stretching vibrations, amide carbonyl stretching vibrations (Amide
I), and CeNeH deformation vibrations (Amide II) are found at
ꢁ
ꢁ
from ꢂ60 C to 140 C.
The viscoelastic properties of PEAs were tested by a dynamic
mechanical analyzer (DMA), model Q800 (TA Instruments, New
Castle, DA,) equipped with a liquid nitrogen cooling system. The
ꢂ1
ꢂ1
ꢂ1
3300 cm , 1645 cm and 1537 cm , respectively [26,27]. The
ratio of C]O (ester) peak area to the NeH (amide) peak area was
0.62 for PEA (I), 1.27 for PEA (II), and 1.90 for PEA (III), i.e., a 1:2:3
series, consistent with the actual ratio of ester to amide groups in
the PEA (I):PEA (II):PEA (III) structures.
samples
17.5 mm ꢀ 10 mm ꢀ 0.5 mm) and were analyzed in a single
cantilever mode. The samples were heated under a constant rate of
were
cut
into
a
rectangular
shape
(
ꢁ
ꢁ
ꢁ
2
C/min over a temperature range of ꢂ60 C to 40 C. The
measurements were performed following the ASTM E1640-99
3.2. Hydrogen bonding structures of the PEAs
standard at a frequency of 1 Hz and fixed oscillation displace-
ment of 15 mm.
The chemical structure of PEAs repeat units and hypothetical
The measurements of tensile strength and tensile strain were
obtained using a Texture Analyser (TA HD, Texture Technologies
Corp, NJ, U.S.A.) equipped with a 2 kg load cell. Tensile tests were
performed on PEAs processed into films using a simple hot-press
technique. The samples were die cut by an ASTM D638 type V
cutter to determine the mechanical properties of the polymer.
arrangements of the hydrogen bonding structures in the synthe-
sized PEAs with ester to amide ratio of 1:1, 2:1 and 3:1 are shown in
Fig. 3aec, respectively. PEA (I) has a repeating unit with the shortest
chain length (molar mass of 484 g/mol), followed by PEA (II) (molar
mass of 712 g/mol and a chain length w30% longer), then PEA (III)
(
molar mass of 1025 g/mol and a chain length w30% longer than
Tensile strength and elongation were tested at a temperature of
ꢁ
PEA (II)). Due to the existence of one amide group in each structure,
the polymer chains are anchored by two hydrogen bonding sites,
regardless of the chain length, imparting a decreasing hydrogen-
bond density from PEA (I) to PEA (II) to PEA (III). The data related
to the repeating units and the hydrogen bonding of the PEAs are
listed in Table 2.
2
0 C. The sample was stretched at 50 mm/min from a gauge of
5 mm.
3
Crystallinity and crystalline structures were examined by wide
angle X-ray diffraction (WAXD). X-ray diffraction was carried out by
an EMPYREAN diffractometer system (PANalytical, The
Netherlands) equipped with a filtered Cu-K
a radiation source
(
l
¼ 1.540598 Ǻ), and a PIXcel-3D area detector. The scanning range
ꢁ
ꢁ
ꢁ
q) with a step size of 0.026 ,
for PEA samples was from 3.3 to 80 (2
920 points were collected in this process. The application software
3.3. Thermal and phase behavior of PEAs
2
was Data Collector 3.0, and the data were analyzed using X’Pert
HighScore 3.0 software.
Fig. 4a displays the TGA thermogram profiles of the PEAs
obtained under a constant nitrogen flow. Overall, the three PEAs
ꢁ
were found to be thermally stable to around 300 C and to have
3
. Results and discussion
similar complex decomposition behavior as clearly displayed by the
TGA derivative curves (Fig. 4b). However, PEA (III) displayed slightly
higher thermal stability compared to PEA (I) and PEA (II), explain-
able by its relatively higher molecular weight, a parameter known
to affect proportionally the decomposition temperature [28,29].
3.1. Characterization
The chemical structure of the synthesized polymers was
1
13
confirmed using H NMR, C NMR and FTIR spectroscopy. All three
PEAs demonstrated similar number average molecular weight (M ).
n
Table 1
The molecular weights and distribution values of PEAs as deter-
mined by GPC are listed in Table 1.
GPC results of the PEAs (M
weight average molecular weight; PDI ¼ M
n
is the number average molecular weight; M
/M ).
w
is the
PDI
1
w
n
The H NMR characteristic peaks of the amide and ester linkages
M
n
M
w
of the PEAs were observed at 6.37e6.12 and 4.15e4.01 ppm,
respectively (Fig. 1). These two peaks are characteristic of the
hydrogen atom attached to the nitrogen in the amide linkage and
the methylene groups adjacent to the oxygen in the ester linkage,
4
4
4
4
4
4
PEA (I)
PEA (II)
PEA (III)
2.08 ꢀ 10
2.09 ꢀ 10
2.26 ꢀ 10
2.98 ꢀ 10
3.44 ꢀ 10
3.63 ꢀ 10
1.43
1.64
1.61

4508
J. Zuo et al. / Polymer 52 (2011) 4503e4516
Fig. 1. 1H NMR spectra of the polyesteramides with varying ratios of ester to amide linkages. (a) PEA (I): ester:amide ¼ 1:1, (b) PEA (II): ester:amide ¼ 1:2, (c) PEA (III): ester:amide ¼ 1:3.

J. Zuo et al. / Polymer 52 (2011) 4503e4516
4509
The different decomposition stages and the degradation
temperatures of each stage were associated with specific common
decomposition processes of the PEAs. The two initial stages of
decomposition were attributed to the breakage of the ester and
amide linkages, respectively, based on similar degradation studies
on PEA using combined differential thermal gravimetry (DTG) and
FTIR techniques conducted by Sudha and Pillai [30]. A thermal
degradation study comparing the weight loss for polyester amide
copolymers with varying ratios of ester and amide blocks also
suggested that the first decomposition event is attributable to the
ester blocks, and the second to the amide blocks [31].
ꢁ
The DSC thermogram obtained by heating (10 C/min) the
ꢁ
preformed polymer films to 140 C and referred to as the first
ꢁ
heating cycle is shown in Fig. 5a. The second DSC heating (3 C/min)
thermogram obtained after the sample was crystallized from the
ꢁ
ꢁ
ꢁ
melt (140 C) at a rate of 5.0 C/min down to ꢂ60 C then heated
ꢁ
(3
C/min) is shown in Fig. 5b. The apparent glass transition,
melting and melt-crystallization temperature, and enthalpy of
melting (T , T , Tmc, and , respectively) of the PEAs obtained
from the two cycles are listed in Table 3.
Fig. 2. FTIR spectra of the polyester amides with varying ratios of ester to amide
linkages. PEA (I): ester:amide
¼
1:1, PEA (II): ester:amide
¼
1:2, PEA (III):
g
m
DH
m
ester:amide ¼ 1:3.
O
O
O
H N
O
H N
O
H N
O
O
O
O
O
O
O
HN
H N
H N
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
H N
O
H N
O
H N
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
H N
H N
H N
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
HN
H N
H N
O
O
O
O
O
O
O
O
O
O
H N
H N
H N
O
O
O
H N
H N
H N
O
O
O
a
HN
H N
H N
b
O
O
O
H N
H N
H N
c
Fig. 3. Chemical structure of repeat units and hypothetical arrangements of the hydrogen bonding structures in the synthesized PEAs with ester to amide ratio of (a) 1:1; (b) 2:1; (c)
:1.
3

4510
J. Zuo et al. / Polymer 52 (2011) 4503e4516
Table 2
thermodynamically most stable phase (melts at the highest
temperature of 87.8 C). For the same processing conditions, the
Structural and hydrogen-bond information of the PEAs.
ꢁ
PEA (I)
PEA (II)
PEA (II)
polymer chains of PEA (II) and PEA (III), which are longer and have
significantly lower hydrogen bonding density than PEA (I), ther-
modynamically less stable solid phases were formed.
Molar mass of the repeating unit (g/mol)
Number of HB sites per repeating unit
Total HB sites in each polymer
Length of the repeating unit (Å)
Hydrogen-bond density (1/Å)
484
2
86
31
712
2
58
48
0.042
1025
2
44
73
0.028
The lower temperature of melting of the crystals detected in the
ꢁ
preformed PEA (III) compared to those in PEA (II) (49 vs. 55 C) is
0.065
not insignificant e PEA (II) has 30% shorter repeating units,
meaning that PEA (II) chains are relatively easier oriented, allowing
for more stable crystal packing, and of course, PEA (II) has signifi-
cantly denser hydrogen bonding, resulting in more attractive forces
acting in the crystal lattice. Indeed, this trend of course is strongly
supported with the significantly higher melt temperature of the
crystals formed by PEA (I). It should also be noted that the DSC
scans do not indicate that the entire polymer in any case was
crystalline e indeed, since we did not have either a totally amor-
phous polymer or totally crystalline polymer for comparison
purposes, DSC is not useful in ascertaining the degree of crystal-
linity. However, as will be discussed later, WAXD scans of the
polymers studied indicate a significant amount of the polymer
chains existed in amorphous form. Additionally, as is also discussed
later, WAXD measurements support the presence of different type
of crystals in the polymers.
HB: Hydrogen bonding.
The thermograms obtained by heating the preformed samples
displayed a glass transition and three endotherms (Fig. 5a). Note
that beside the prominent endotherm, PEA (I) showed also two
very small and well-resolved endotherms (not evident but indi-
cated by arrows in Fig. 5a and listed in Table 3). The first endo-
thermic event observed in PEA (II) and PEA (III) showed as a leading
shoulder to the more important and well-resolved peak, indicating
the overlap of two melting events. Obviously, two solid phases
having different crystal stabilities were involved. Note that the peak
maxima were well located using the first and second derivative of
the heat flow signal, but only their combined area has been
measured (values listed in Table 4). PEA (I) did not show any exo-
therm, whereas, PEA (II) and PEA (III) displayed a broad and intense
exotherm following the two first unresolved endotherms.
The second DSC heating cycle shown in Fig. 5b further supports
the arguments proffered above to explain the first melting cycle.
The recording of an exotherm in both PEA (II) and PEA (III) is
characteristic of melt crystallization. As the temperature was
increased, their polymer chains acquired more mobility, recrystal-
lized from the melt into their most stable form possible then
subsequently melted.
ꢁ
The melted polymer films in this case were cooled at a rate of 5 C/
min, a relatively much faster cooling rate than the ambient cooling
for the formation of the polymer films the first time. Interestingly,
at this faster rate, even the relatively short repeating units in PEA (I)
are unable to orient before the glass transition temperature is
Note that the enthalpy of melting recorded for the combined
first two peaks of PEA (I) and PEA (II) was 30 J/g, practically the
same value of the total enthalpy recorded in the melting of PEA (I).
This low value can be linked to the relatively low crystallinity
achieved in these samples as will be discussed in the crystalline
structure section. This constant value of the enthalpy of melting
irrespective of the repeating unit length indicates the crucial
importance of the total hydrogen bonding in the polymorphism of
the PEAs, for the preformed polymers. However, the recording of an
overlap between these endotherms and the following exotherm
indicated that the enthalpy of melting of the second phase was not
totally recorded, alluding to other important contributions to the
polymorphism of the PEA, such as the van der Waals forces.
The differences observed in the transformation paths are
explainable by the relative size of the repeating units and hydrogen
bonding density. The cooling at room temperature of the pre-
formed polymers was probably slow enough to allow for the
shorter chains with the densest hydrogen bonding structure of PEA
reached, but as the polymer is melted, above the T , chains acquire
g
enough molecular mobility to orient and a burst of crystallization is
experienced, these crystals then immediately remelting at the melt
ꢁ
point of 87.2 C as before. This behavior is seen also with PEA (II),
with a burst of crystallization occurring above the T , although
g
again because of the relatively longer chain segments and because
of the smaller density of hydrogen bonding sites, the types of
crystals formed through the relatively limited mobility of the
chains above T are weaker, less organized crystals compared to the
g
most thermodynamically stable crystal type possible to be formed
by this polymer. These less stable crystals melt, and mediate
a second crystallization event, the most thermodynamically stable
ꢁ
crystals then finally melt at 76.2 C. The effect of lower density of
hydrogen bonding and longer repeating segments is exacerbated in
PEA (III) as evidenced by the very small crystallization event which
occurs above the T (indeed the extent of the crystallization event
g
above the T steadily decreases from PEA (I) to PEA (II) to PEA (III)).
g
(
I) to pack preponderantly in
a
relatively well-developed,
As with PEA (II), the relatively weak crystals formed are melted
ꢁ
Fig. 4. TGA (a) and TGA derivative (b) curves of PEAs at a heating rate of 10 C/min.

J. Zuo et al. / Polymer 52 (2011) 4503e4516
4511
ꢁ
ꢁ
Fig. 5. DSC heating thermograms of the PEAs (a) First heating cycle obtained by heating (10 C/min) the preformed polymers. (b) Second heating cycle obtained by heating (3 C/
min) a sample crystallized (5 C/min) from the melt.
ꢁ
immediately after the initial crystallization event, which mediates
a sharp crystallization event of the most stable crystals, which melt,
together with the finite amount of thermodynamically most stable
forms which may have formed during cooling, at the temperature
(see Fig. 3) requiring lower temperatures to initiate the movement
of the chain segments, thus yielding lower T . It is also worth noting
g
that the hydrogen bonding is not only limited to the amide-ester
groups shown in Fig. 3, but may also extend to ester groups adja-
cent to the primary ester groups indicated in the figure. This would
result of course in some heterogeneity in the degree of hydrogen
bonding, as the polymer chains themselves are also mobile,
bringing adjacent ester groups closer or further away from the
amide linkages. This heterogeneity will further contribute to
ꢁ
of 74.3 C as before. The shifts in the melting peak maximum of the
most stable thermodynamic forms of crystals for the three PEAs
between the first and the second heating cycles is not unexpected e
the melting events are broad and there is an annealing effect caused
by successive melting and crystallization, resulting in slight
changes to packing. However, the relative differences between
these events are similar, so that the analysis can be with confidence
conducted in the manner it was above.
g
a deviation from linearity of the dependence of T on the primary
hydrogen-bond density as presented in Fig. 6. However, the effect of
the primary (as drawn in Fig. 3) hydrogen-bond density was
predominant in this study as indicated by the approximately linear
The T
displayed in Fig. 6. The PEA type is reported in the top axis of the
figure. As can be seen, T was highest for PEA (I) followed by PEA (II)
then PEA (III). The relatively large drops in T as the ester to amide
g
obtained by DSC versus hydrogen bonding density is
g
change of T vs. hydrogen-bond density displayed in Fig. 6. The
g
above structural considerations also of course extend to the
mechanical properties of the polymers, and therefore although in
succeeding sections only the predominant effects of the primary
hydrogen-bond density is discussed in detail, it is meant to be
understood that there is a certain level of heterogeneity in the
g
ratio was increased can be well explained by the hydrogen bonding
structure of the polymers. The intermolecular hydrogen bonds
present in PEA, due to the amide groups, contribute significantly to
preventing the mobility of the polymer chains. As the chain length
between amide groups increases, hydrogen-bond density
decreases from its highest value in PEA (I) to its lowest in PEA (III)
hydrogen bonding of the polymer chains. The information of T
change in relation to the ester to amide ratio can help in the design
of similar structures derived from oleic acid, with targeted T values
according to the required application conditions.
g
g
Multiple melting behavior of polymers, related to the formation
of different polymorphs due to limited mobility of the long poly-
meric chains have been noted by numerous studies, and in
particular for polyesters and polyester amides [31e34]. Although
the thermal stability was not affected by the ester to amide ratio
from the thermal decomposition perspective, it was affected at the
melting level e with the crystals formed by PEA (I) being more
thermally stable than those formed by PEA (II), which in turn are
more stable than those formed by PEA (III). This motivates the
postulate that the thermal behavior of the PEAs is indeed signifi-
Table 3
Melting values obtained from the heating cycles of the PEAs. The peak temperature
of the endotherm and exotherm (T
highest to the lowest temperature. Enthalpy (in J/g) is the area under the peak. The
values obtained from DMA measurements (Peak of tan ) are listed for comparison
purposes.
m
and Tmc, respectively) are numbered from the
T
g
d
1
st heating cycle
2nd heating cycle
ꢁ
T ( C)
T
m1
T
mc
T
m2
T
m3
T
m1
T
mc1
T
m2
T
mc2
PEA (I)
87.8
25
77.4
44
76.0
50
NA
NA
56.0
30
58.8
32
64.9
1
45.3
4
87.2
33
66.1
33
NA
NA
D
H (J/g)
PEA (II)
H (J/g)
PEA (III)
H (J/g)
49.3
29
40.0
76.2
48
52.5
31
48.0
9
14.2
24
g
cantly affected, but only at the T and the crystalline melting levels,
i.e., phase development, not at the degradation of the polymer
D
54.9
30
41.0
74.2
56
55.8
32
50.4
22
31.9
1.5
D
ꢁ
Table 4
T
g
( C)
Mechanical properties of the PEAs obtained from tensile analysis.
DSC
DMA
Ultimate strength (MPa) Maximum strain (%) Young’s modulus (MPa)
PEA (I)
PEA (II)
PEA (III)
3.4 ꢃ 0.8
ꢂ20.0 ꢃ 0.3
ꢂ34.1 ꢃ 0.1
17.9 ꢃ 0.3
ꢂ1.6 ꢃ 0.1
ꢂ15.0 ꢃ 0.1
PEA (I) 19.6 ꢃ 0.3
PEA (II) 10.2 ꢃ 0.1
PEA (III) 8.5 ꢃ 0.1
12.4 ꢃ 0.8
14.3 ꢃ 0.3
11.8 ꢃ 0.8
586 ꢃ 7
310 ꢃ 3
234 ꢃ 4
NA: not applicable.

4512
J. Zuo et al. / Polymer 52 (2011) 4503e4516
PEA (I)
PEA (II)
PEA (III)
required to deform the sample is elastically recovered instead of
2
0
0
0
being dissipated as heat [35]. It is clearly shown in Fig. 8 that at
ꢁ
room temperature (around 20 C), tan
d decreases from PEA (I) to
1
PEA (III), indicating that the film of PEA (III) had the highest rubbery
elasticity, and the film of PEA (I) recorded the lowest.
The T
g
was also investigated by DMA under a frequency of 1 Hz.
are listed along the T
g
T
g
values reported as the peak value of tan
d
-
-
-
-
-
10
20
30
40
50
values obtained by DSC in Table 3. Because DSC measures the
changes of heat capacity during the glass transition, whereas, DMA
measures the mechanical response of the polymer chains during
the transition; the respective recorded values for T
g
value are
different. The trend is however the same.
Mechanical properties were also examined using tensile anal-
ysis. The tensile stress versus strain curves of the PEAs are displayed
in Fig. 9. The slope of the linear segment in each curve (shown as
extended straight lines in Fig. 9) is used to determine Young’s
modulus. The ultimate tensile strength, maximum strain and
Young’s modulus for PEAs are listed in Table 4. PEA (I) has the
highest ultimate tensile strength (19.6 MPa) followed by PEA (II)
0
.07
0.06
0.05
0.04
0.03
0.02
Hydrogen-bond density (1/Å)
Fig. 6. T
g
value obtained by DSC as a function of hydrogen-bond density. The ester to
amide ratio is reported on the top axis. The dashed line is a fit to a line (R ¼ 1.000).
2
(
(
10.2 MPa) then PEA (III) with the lowest ultimate tensile strength
8.5 MPa).
The difference in ultimate tensile strength (Fig. 10) between PEA
I) and PEA (II) is much larger (9.4 MPa) than the difference
between PEA (II) and PEA (III) (1.7 MPa), outlining the dramatic
effect of hydrogen bonding density on tensile properties of the
PEAs, in accordance to what has been discussed earlier to explain
the dynamic mechanical properties of the PEAs.
linkages. That is, it seems that the degradation of the ester and
amide linkages are unaffected by the fact that they are both present
in a single polymeric chain, at least in these polymers.
(
3.4. Mechanical properties of the PEAs
Young’s modulus (Table 4) was determined by the slope of the
linear segment in the stress versus strain curve. The Young’s
modulus decreased dramatically with hydrogen-bond density,
Fig. 7 displays the storage moduli of PEAs as a function of
temperature. It was shown that the storage modulus decreased as
the ester to amide ratio changed from 1:1 to 2:1 and to 3:1. It was
(Fig. 11) indicating a monotonous dependence on hydrogen
also worth noting that the stored energy at room temperature
bonding. The addition of a first amide block decreased the value of
Young’s modulus by 47%, whereas the second decreased it by
a further 25%, which if the uncertainty attached to the measure-
ments is taken into account, is consistent with the amide ratio 1:2:3
of the PEAs. This trend can help to design oleic acid-derived PEA
structures with targeted mechanical properties.
Fig. 9 also shows that the elongation at break is statistically
similar for PEA (I) and PEA (III), but it is higher for PEA (II). At this
time, we can find no structural reason for this noted difference.
ꢁ
(
around 20 C) was largely reduced from PEA (I) to PEA (II), but the
reduction was much smaller from PEA (II) to PEA (III). As explained
above, PEA (I) had approximately 86 hydrogen-bond sites in a linear
chain, whilst PEA (II) had 58 and PEA (III) had 44. Clearly, there was
a large decrease in the density of hydrogen bonding between PEA
I) and PEA (II), but not as drastic a decrease between PEA (II) and
PEA (III). This trend of mechanical properties was further confirmed
(
by the results of tensile analysis.
The value of tan
to investigate elasticity of the polymer. Because tan
loss modulus to storage modulus, it is the ratio of viscous to elastic
d
(Fig. 8) may also be analyzed for each PEA film
d
is the ratio of
3.5. Crystalline structure
0
0
0
components of the modulus (tan
d
¼ E /E ). The smaller tan
d is, the
The crystalline structure of the PEA samples was investigated by
wide angle X-ray diffraction (WAXD). The WAXD patterns of PEA (I),
0
00
larger the E is relative to E , which means more of the energy
Fig. 7. Storage modulus as a function of temperature measured by DMA for PEAs.
Fig. 8. Tan d as a function of temperature measured by DMA for PEAs.

J. Zuo et al. / Polymer 52 (2011) 4503e4516
4513
PEA (I)
PEA (II)
PEA (III)
700
600
500
400
300
200
0.07
0.06
0.05
0.04
0.03
0.02
Hydrogen-bond density (1/Å)
Fig. 9. Tensile stress versus strain curves of PEAs. The straight lines are extensions of
the linear segment in each curve.
Fig. 11. Young’s modulus of PEAs as a function of hydrogen-bond density. Sample type
is reported on the top axis.
(
II) and (III) shown in Fig. 12, suggest similar but complex poly-
morphic structures. The WAXD patterns present a large amorphous
halo and resolved diffraction peaks indicative of the semicrystalline
nature of the samples. It is well known that the X-ray scattering
pattern of an amorphous polymer contains one or more halos
corresponding to van der Waals spacings or larger [36]. As
customarily done in the case of semicrystalline polymers, the
amorphous contribution was fitted with a linear combination of
two amorphous profiles and were located at the diffraction angles
The observed intensities of the crystalline phases were evalu-
ated by integrating the crystalline peaks observed in the X-ray
diffraction profiles, after the subtraction of the background and
amorphous contributions. The relative crystallinity was obtained
from the integrated intensity over the observed Bragg reflections
normalized by that over the whole WAXS profile. Results of WAXD
data analysis are summarized in Table 5.
Under the right conditions, almost all of the semicrystalline
polymers can form different polymorphs. Because the thermal and
mechanical studies indicated a very strong effect of the hydrogen
bondings, we will discuss the crystal structure of the PEAs in light of
the polymorphism reported for the polyamides. Note that
ꢁ
ꢁ
(
2
q) of approximately 10.5 and 19.3 , similarly to what has been
reported in several papers (see for example [37e39]).
The presence of halos indicates the existence of a degree of
packing regularity in amorphous polymers that should be consid-
ered in understanding of polymer chain conformation [40]. As
usually done for similar polymers which do not have any side
chains, the amorphous profiles were attributed to interchain
distances in the amorphous phase (damorph) which were calculated
using the relation [41]:
damorph ¼ 1:11 ꢀ dBragg
where dBragg is the largest Bragg d-spacing of the two amorphous
profiles.
PEA (I)
PEA (II)
PEA (III)
24
20
16
12
8
4
0.07
0.06
0.05
0.04
0.03
0.02
Hydrogen-bond density (1/Å)
Fig. 12. Wide angle X-ray diffraction patterns of preformed PEA films measured at
Fig. 10. Ultimate strength of PEAs in this study as a function of hydrogen-bond density.
Sample type is reported on the top axis.
room temperature. (a) PEA (I) with ester:amide
ester:amide ¼ 2:1; (c) PEA (III) with ester:amide ¼ 3:1.
¼
1:1; (b) PEA (II) with

4514
J. Zuo et al. / Polymer 52 (2011) 4503e4516
similarities can be drawn from the structures reported for other
crystalline polymers such as particularly polyesters [42e45] or poly
where A(i) is the area of the characteristic peaks of i ¼
a, g and
hexagonal phase. The results are listed in Table 5.
(
vinylidene fluoride) [46e49]. However, because there is no
common labeling scheme for the different forms observed in the
semi)crystalline polymers, a comparative analysis, which is not the
Taking into account the uncertainties attached to the calcula-
tions, one can notice that the estimated amount of -phase was
the same in the three PEAs. The -phase increased from 33% in
a
(
g
topic of our report, would be onerous. The WAXD spectra were
therefore primarily analyzed in light of the very extensively studied
structures observed in nylon, a prominent member of the poly-
amide class of semicrystalline polymers [50].
PEA (I) to w53% in both PEA (II) and PEA (III), whereas, the
hexagonal phase decreased from 40 to w25%. Clearly, the lower
hydrogen-bond density due to the addition of one amide block
promoted the “twisted” phase at the detriment of the less stable
hexagonal phase, whereas, a further addition of an amide block
did not. The polymers’ processing conditions have obviously
limited the effect of hydrogen bonding distribution on further
phase development. This limitation can be linked to size and mass
effects. However, with increasing ester to amide ratio, a decrease
in the width of the characteristic lines of the three phases was
observed, indicating an increase in crystalline homogeneity and
perfection. This ordering effect is therefore attributable to the
ester blocks.
Two sets of WAXD lines corresponding to two conformations
similar to those observed in the case of Nylon have been isolated.
ꢁ
The first set is constituted of the 20.1, 23.2 and 24.7 lines (4.41, 3.84
and 3.64 Å, respectively) and the second set 16.9 and 19.1 (5.24 and
4
a
.64 Å, respectively). In addition to the peaks characteristic of the
ꢁ
and
g
crystals, there remains a diffraction peak at 2
q
¼ 21.18 ,
with a spacing of 4.2 Å related to another polymorph formed in the
PEA samples.
The first set (4.41, 3.84 and 3.64 Å) corresponds to room
temperature spacings from the (100), (010)/(110) and (110)
Also, the ester groups of the anti-parallel chains within the unit
reflections of the
sponds to the spacings from the (121), and (010) reflections of the
so-called -form, both of which are well known to form in nylon
40,51,52]. Note that the structure is also well known for the n-
alkanes [53]. The -phase has monoclinic structure, contains
zigezag chains with the hydrogen bonds situated between anti-
parallel chains, whereas, the -phase although also having
a
-form and the second (5.24 and 4.64 Å) corre-
cell play an important role the formation of the
gested by conformational analysis based on intramolecular
potential energy calculations of poly- -hydroxybutyrate [58].
Similarly to this polymer, the ester groups in the PEAs may be
nearly at the same level allowing for the dipoleedipole interaction
to play a balancing role in the development of the molecular
g-form as sug-
g
b
[
a
a
g
packing of the PEAs. It is therefore understandable that the g-
a monoclinic structure, contains helical chains allowing hydrogen
bonds to be formed between parallel chains [54]. The main struc-
phase content increase was not only due to the decrease in
hydrogen bonding density but also that the electrostatic dipo-
leedipole interactions between the dipoles associated to the ester
groups contribute considerably to the stabilization of the helical
tural difference between the
a- and
g
-crystals come from the
-form in fact repre-
molecular packing within the unit cell. The
a
sents a variation from the helical to the planar (or nearly planar)
zigezag type of chains [55].
conformation of
[59] in the case of poly(S-lactic acid). The relatively higher width
displayed by the characteristic lines of the -form indicates
significant disorder both along and perpendicular to the chain axis
similarly to what has been observed in several aliphatic polyesters
such as PCL [60].
g-form crystals as argued by De Santis and Kovacs
The diffraction peak at a spacing of 4.2 Å represents probably
a hexagonal structure, which has also been reported for this type of
material [50,56]. Note that these phases are known to possibly
coexist with relative content of the phases depending on the pro-
cessing conditions [57].
a
The structural features of the a-, g- and hexagonal forms and
Quantitative evaluations of the relative content of the crystalline
forms possibly present in the samples were estimated from the
their relative amounts in the samples can explain the phase
developement displayed by the DSC heating profiles of the different
PEAs. The recorded thermal events (endotherms and recrystalli-
zation) are obviously related to the melting and further phase
ꢁ
ꢁ
2
q
¼ 15 e30 region of the WAXD patterns. The percent content of
the different forms in the crystalline fraction was estimated using
the relation:
change of the preformed solid phases (solid
form). Phase change in polymeric materials such as our PEAs is very
complex. The -phase is stable and may be converted into the
phase by melting and recrystallization [61] and the hexagonal
phase can be converted into the -phase by annealing [51]. The
a and g, and hexagonal
AðiÞ
g
a-
%
Phase ðiÞ ¼ 100 ꢀ P
AðiÞ
i
a
similar and relatively low crystallinity (w50%) estimated from the
WAXD patterns of the PEA can be related to the similar and low
enthalpy of melting recorded for the melting of the solid phase
Table 5
Structural data obtained from the WAXD spectra of the PEAs: dhkl is Bragg distance
with the indices obtained by comparing the experimental reflection to similar forms
already formed. The increase of (
g
þ
a) at the detriment of the
observed in polyamides and polyesters. P
a g H
, P , and P are the relative phase contents
hexagonal phase content as the ester to amide ratio increases can
explain the differences in thermal behavior displayed by the
samples as discussed in the melting behavior section.
of the -, -, and hexagonal phases, respectively. damorph is the interchain distance in
a
g
the amorphous phase. Calculated uncertainties are w5% as determined by the
standard error obtained for three replicates.
a-form
g-form
Hexagonal form
3
3
.6. Comparison of PEAs
d (Å)
(hkl)
d (Å)
(hkl)
d (Å)
(hkl)
(100)
4
3
3
.4
.8
.6
(100)
(010)/(110)
(110)
5.2
4.6
(121)
(010)
4.2
.6.1. Comparison with lipid-based PEAs
Compared to other lipid-based PEAs [20,21,62,63], the polymers
prepared in this study displayed superior decomposition stability
and lower T , indicating that they can be used over a wider range of
X
cryst ꢃ 5 (%)
damorph ꢃ 0.5 (Å)
Relative phase
content (%)
g
temperature. The above-cited studies did not report any mechan-
ical property value or even mention if films were made or tested. As
expected, the melting temperatures reported for PEAs presenting
aromatic groups [21,63] were higher than those measured in PEA (I)
to (III).
P
a
P
g
P
H
PEA (I)
PEA (II)
PEA (III)
45
50
52
10.7
10.8
10.3
33
54
52
27
20
24
40
26
24

J. Zuo et al. / Polymer 52 (2011) 4503e4516
4515
3.6.2. Comparison with petroleum-based PEAs
Acknowledgements
The T and T of PEA (I)e(III) are well in the range of the values
g
m
reported for petroleum-based PEAs and tensile strength is
comparable [1,7,27,64e69]. The elongation measured in PEA
We express our thanks to the Ontario Soybean Growers/Grain
Farmers of Ontario, Elevance Renewable Sciences, NSERC, GPA-EDC,
Industry Canada, and Trent University for their financial support for
the research.
(
I)e(III) is rather smaller than most of reported values, but the
elongation does not pose a problem if these PEAs were to be used as
coating materials. The decomposition temperatures measured for
ꢁ
ꢁ
our PEAs (w300 C) are relatively lower compared to the w340 C
References
ꢁ
reported by Lips et al. [27] or 380 C by Amelin et al. [69], but are
better compared to the 230e280 C values reported in Refs.
ꢁ
[
[
[
[
[
[
[
[
[
1] Jokhadze G, Machaidze M, Panosyan H, Chu CC, Katsarava R. Journal of
Biomaterials Science-Polymer Edition 2007;18(4):411e38.
2] Katsarava R, Beridze V, Arabuli N, Kharadze D, Chu CC, Won CY. Journal of
Polymer Science Part A: Polymer Chemistry 1999;37(4):391e407.
[64,65,68].
4
. Conclusions
3] Vera M, Almontassir A, Rodriguez-Galan A, Puiggali J. Macromolecules 2003;
36(26):9784e96.
4] Tsitlanadze G, Machaidze M, Kviria T, Djavakhishvili N, Chu CC, Katsarava R.
Journal of Biomaterials Science-Polymer Edition 2004;15(1):1e24.
5] Legashvili I, Nepharidze N, Katsarava R, Sannigrahi B, Khan IM. Journal of
Biomaterials Science-Polymer Edition 2007;18(6):673e85.
6] Lee SH, Szinai I, Carpenter K, Katsarava R, Jokhadze G, Chu CC, et al. Coronary
Artery Disease 2002;13(4):237e41.
7] Li L, Chu CC. Journal of Biomaterials Science-Polymer Edition 2009;20(3):
341e61.
8] Oledzka E, Narine SS. Journal of Applied Polymer Science 2011;119(4):
It was shown by TGA analysis that the thermal stability of the
PEAs is not significantly influenced by the ratio of the ester to amide
linkages. However, the melting point of PEAs decreased from PEA
(
I) to PEA (III), indicating a correlation to the decrease of ester to
amide ratio.
g
The T values measured by DSC and DMA revealed that the T is
g
largely affected by the hydrogen-bond density of the polymer. As
the ester to amide ratio increased from 1/1 to 3/1, the hydrogen
1873e82.
9] Oledzka E, Kong X, Narine SS. Journal of Applied Polymer Science 2011;
bonding sites in the polymer chains decrease and so does the T
Mechanical properties were also significantly influenced by the
hydrogen-bond density of the PEAs. tan decreases from PEA (I) to
g
.
119(3):1848e56.
[
[
10] Narine SS, Kong XH, Bouzidi L, Sporns P. Journal of the American Oil Chemists
Society 2007;84(1):55e63.
11] Narine SS, Kong XH, Bouzidi L, Sporns P. Journal of the American Oil Chemists
Society 2007;84(1):65e72.
d
PEA (III) at room temperature, indicating that elasticity of the
polymer film increases from PEA (I) to PEA (III).
[12] Kong XH, Tan SS, Narine SS. Journal of Applied Polymer Science 2009;114(1):
39e48.
1
The storage modulus measured using DMA showed that the
stored energy in PEAs decreased from PEA (I) to PEA (III). The
tensile analysis also indicated that the ultimate tensile strength and
the Young’s modulus of the PEA were reduced when the hydrogen
bonding decreases. However, the effects of ester to amide ratio on
mechanical properties did not change linearly, due to the non-
linear decrease of the hydrogen-bond density of PEAs. The
hydrogen-bond density had a significant change between PEA (I)
and PEA (II), and the change decreases between PEA (II) and PEA
[
[
13] Kong XH, Narine SS. Biomacromolecules 2008;9(8):2221e9.
14] Dumont MJ, Narine SS. Journal of Applied Polymer Science 2010;118(6):
3211e7.
[
[
15] Hojabri L, Kong XH, Narine SS. Journal of Polymer Science Part A: Polymer
Chemistry 2010;48(15):3302e10.
16] Biermann U, Bornscheuer U, Meier MAR, Metzger JO, Schäfer HJ. Angewandte
Chemie International Edition 2011;50(17):3854e71.
[
[
[
17] Gandini A. Green Chemistry 2011;13(5):1061e83.
18] Mahapatra SS, Karak N. Progress in Organic Coatings 2004;51(2):103e8.
19] Dutta S, Karak N. Progress in Organic Coatings 2005;53(2):147e52.
[20] Ahmad S, Ashraf SM, Naqvi F, Yadav S, Hasnat A. Progress in Organic Coatings
003;47(2):95e102.
21] Li XR, Su YL, Chen QY, Lin Y, Tong YJ, Li YS. Biomacromolecules 2005;6(6):
181e8.
2
(
III).
The crystallinity of the various PEAs was a function of their
[
3
structure, particularly repeating unit length and density of
hydrogen bonding sites. Wide angle X-ray diffraction confirmed the
nature of the crystalline structures of the PEAs, The characteristic
peaks presented in the X-ray patterns revealed that two forms of
[22] Hill JW, McEwen WL. Organic Syntheses 1943;2:53e4.
[23] Hojabri L, Kong XH, Narine SS. Biomacromolecules 2010;11(4):911e8.
[
24] Kondo F, Kakimi S, Kimura H, Takeishi M. Polymer International 1998;46(4):
39e44.
3
[25] Bouzidi L, Boodhoo M, Humphrey KL, Narine SS. Thermochimica Acta 2005;
439(1e2):94e102.
crystalline structures were formed in each polymer e
a form and
[
[
26] Sava I. Revue Roumaine De Chimie 2009;54(5):381e7.
27] Lips PAM, Broos R, van Heeringen MJM, Dijkstra P, Feijen J. Polymer 2005;
b
form. The form which represents a helical conformation was
a
increased from PEA (I) to PEA (III). This is due to the fact the PEA (I)
has the highest hydrogen-bond density, and it is relatively harder
for the polymeric chains to rotate to the helix structures.
The polymers all demonstrated a degree of crystallinity which is
strongly influenced by the kinetics of cooling, particularly as evi-
denced by melt-mediated transformations and by the shifting of
melting peaks on cycling. This suggests that more work should be
done on elucidating the various local thermodynamic minima that
are presented in these structures, constrained by the kinetics of the
system, and investigation of the relative time-stability of these
states and their concomitant physical functionalities. It is quite
likely that significant physical functionality changes may be effec-
ted through a modification of the degree and type of crystallinity
developed in the films, as a result of varying cooling kinetics.
Comparison of the PEAs reported in the literature suggests that
the properties of PEAs in this study are superior to all other lipid-
based PEAs and comparable with the petroleum-based PEAs. In
addition, the relationships established between mechanical and
thermal properties and structure can significantly help to predict
thermo-mechanical properties of PEAs designed from similar
structures.
4
6(19):7823e33.
[28] Nunes RW, Martin JR, Johnson JF. Polymer Engineering and Science 1982;
2(4):205e28.
29] Wang TL, Hsieh TH. Polymer Degradation and Stability 1997;55(1):
5e102.
[30] Sudha JD, Pillai CKS. Journal of Polymer Science Part A: Polymer Chemistry
003;41(2):335e46.
31] Qian ZY, Li S, He Y, Li C, Liu XB. Polymer Degradation and Stability 2003;81(2):
79e86.
[32] Serrano PJM, Thuss E, Gaymans RJ. Polymer 1997;38(15):3893e902.
2
[
9
2
[
2
[
33] Qian ZY, Li S, Zhang HL, Liu XB. Colloid and Polymer Science 2003;281(9):
69e75.
8
[
34] Armelin E, Paracuellos N, Rodriguez-Galan A, Puiggali J. Polymer 2001;42(19):
7923e32.
[35] Menczel JD, Prime RB. Thermal analysis of polymers: fundamentals and
applications. Hoboken, NJ: John Wiley and Sons; 2009. pp. 398.
36] Miller RL, Boyer RF. Journal of Polymer Science: Polymer Physics 1984;22(12):
[
2043e50.
[37] Guerra G, Vitagliano VM, De Rosa C, Petraccone V, Corradini P. Macromole-
cules 1990;23(5):1539e44.
38] Sun YS, Woo EM. Polymer 2001;42(5):2241e5.
39] Ouchi T, Nagasaka S, Hotta A. Macromolecules 2011;44(7):2112e9.
[40] Murthy NS, Wang Z-G, Hsiao BS. Macromolecules 1999;32(17):5594e9.
41] Alexander LE. X-ray diffraction procedures. New York: Wiley; 1974.
42] Pan P, Inoue Y. Progress in Polymer Science 2009;34(7):605e40.
43] Hoogsteen W, Postema AR, Pennings AJ, Ten Brinke G, Zugenmaier P.
Macromolecules 1990;23(2):634e42.
[
[
[
[
[

4516
J. Zuo et al. / Polymer 52 (2011) 4503e4516
[
[
[
[
[
[
44] Sawai D, Takahashi K, Imamura T, Nakamura K, Kanamoto T, Hyon S-H.
Journal of Polymer Science Part B: Polymer Physics 2002;40(1):95e104.
45] Orts WJ, Marchessault RH, Bluhm TL, Hamer GK. Macromolecules 1990;
[57] Murthy NS, Aharoni SM, Szollosi AB. Journal of Polymer Science: Polymer
Physics Edition 1985;23(12):2549e65.
[58] Cornibert J, Marchessault RH. Journal of Molecular Biology 1972;71:735e56.
[59] De Santis P, Kovacs AJ. Biopolymers 1968;6(3):299e306.
[60] Chatani Y, Okita Y, Tadokoro H, Yamashita Y. Polymer Journal 1970;1:555e62.
[61] Aharoni SM. Crystallinity and polymorphism in the n-nylon family. n-nylons:
their synthesis, structure and properties. Chichester, West Sussex: John Wiley
& Sons, 1997. pp. 34e70.
[62] Zafar F, Ashraf SM, Ahmad S. Progress in Organic Coatings 2004;51(3):
250e6.
[63] Bharathi NP, Khan NU, Shreaz S, Hashmi AA. Journal of Inorganic and
Organometallic Polymers and Materials 2009;19(4):558e65.
[64] Lecomte HA, Liggat JJ, Curtis ASG. Journal of Polymer Science Part A: Polymer
Chemistry 2006;44(6):1785e95.
[65] Arvanitoyannis I, Nakayama A, Kawasaki N, Yamamoto N. Polymer 1995;
36(4):857e66.
[66] De Ruijter C, Bos J, Boerstoel H, Dingemans TJ. Journal of Polymer Science Part
A: Polymer Chemistry 2008;46(19):6565e74.
[67] Okada M, Yamada M, Yokoe M, Aoi K. Journal of Applied Polymer Science
2001;81(11):2721e34.
[68] Harashina H, Nakane T, Itoh T. Journal of Polymer Science Part A: Polymer
Chemistry 2007;45(11):2184e93.
23(26):5368e70.
46] Lovinger A. Developments in crystalline polymers e l. In: DC B, editor.
Developments series. London, UK: Applied Science Publishers; 1982. p. 279.
47] Tashiro K, Tadokoro H. Encyclopedia of polymer science and engineering. New
York: Wiley-Interscience; 1989. pp. 187.
48] Hasegawa R, Takahashi Y, Chatani Y, Tadokoro H. Polymer Journal 1972;3(6):
6
00e10.
49] Farmer BL, Hopfinger AJ, Lando JB. Journal of Applied Physics 1972;43(11):
293e303.
4
[
[
50] Ramesh C, Gowd EB. Macromolecules 2001;34(10):3308e13.
51] Feldman AY, Wachtel E, Vaughan GBM, Weinberg A, Marom G. Macromole-
cules 2006;39(13):4455e9.
[
[
[
[
[
52] Yoshioka Y, Tashiro K. The Journal of Physical Chemistry B 2003;107(43):
1
1835e42.
53] Sirota EB, King JHE, Singer DM, Shao HH. The Journal of Chemical Physics
993;98(7):5809e24.
54] Arimoto H, Ishibashi M, Hirai M, Chatani Y. Journal of Polymer Science Part A:
General Papers 1965;3(1):317e26.
55] Fuller CS, Froch CJ. Further investigations of the chain structure of linear
polyesters. In: Fifteenth colloid symposium. Cambridge, Massachusetts: 1938.
56] Sandeman I, Keller A. Journal of Polymer Science 1956;19(93):401e35.
1
[69] Armelin E, Franco L, Rodriguez-Galan A, Puiggali J. Macromolecular Chemistry
and Physics 2002;203(1):48e58.