J Therm Anal Calorim (2017) 130:919–933

DOI 10.1007/s10973-017-6459-1

Understanding intrinsic plasticizer in vegetable oil-based

polyurethane elastomer as enhanced biomaterial
Ahmad Syafiq Ahmad Hazmi1,2 • Nik Nurfatmah Pz Nik Pauzi1 • Zulina Abd. Maurad1 •
Luqman Chuah Abdullah3,2 • Min Min Aung4 • Azizan Ahmad5 • Mek Zah Salleh6 •
Rida Tajau6 • Mohd Hilmi Mahmood6 • Syahrina Elliyana Saniman7

Received: 31 May 2016 / Accepted: 15 May 2017 / Published online: 6 June 2017
 Akadémiai Kiadó, Budapest, Hungary 2017

Abstract Renewable polyol is of increasing interest as a
building block in biomedical elastomer for bearing
biodegradable ester group and immaculate functionality.
Derived from non-edible vegetable oil, a new class of
elastomer was successfully functionalized with MDI and
TDI. Crosslink densities were varied by regulating ratio of
hydroxyl to diisocyanate (r) at 1/1.0, 1/1.1, and 1/1.2.
Produced elastomers were examined by crosslink density,
attenuated total reflectance Fourier transform infrared
spectroscopy, differential scanning calorimetry, thermo-
gravimetric analysis, dynamic mechanical analysis, tensile
testing, and scanning electron microscopy. The obtained
elastomers had subambient glass transition temperature
(Tg) suggested majority soft segment that acted as a con-
tinuous phase with intermediate phase separation. Medium
conversion at gel point had enhanced physical properties.
Highly elastic mechanical behavior was afforded from
combination of side chains and high molecular weight
polyol. At r = 1/1.2, MDI-based elastomer showed two-
fold improvement in Young modulus at slight expense of
elongation. TDI-based elastomer accomplished elongation
beyond 162%. Branching allophanate and biuret resisted
early thermal breakdown by elevating activation energy.
Frequency response and kinetic of thermal degradation
provided beneficial perspective for elastomer characteri-
zation. The vegetable oil-based polyurethane was found
able to resemble most of the physical properties of poly-
caprolactone (PCL)-derived polyurethane.

& Ahmad Syafiq Ahmad Hazmi 4

Department of Chemistry, Faculty of Science, Universiti
ahmadsyafiq87@gmail.com Putra Malaysia, 43400 Serdang, Selangor, Malaysia
5
1 School of Chemical Science and Food Technology, Faculty
Process Engineering and Design Unit, Advanced
of Science and Technology, Universiti Kebangsaan Malaysia,
Oleochemical Technology Division, Malaysian Palm Oil
43600 Bangi, Selangor, Malaysia
Board, No. 6, Persiaran Institusi, Bandar Baru Bangi,
6
43000 Kajang, Selangor, Malaysia Radiation Processing Technology Division, Malaysian
2 Nuclear Agency, 43000 Bangi, Selangor, Malaysia
Department of Chemical and Environmental Engineering,
7
Faculty of Engineering, Universiti Putra Malaysia, Technology Medical Associate, 40400 Shah Alam, Selangor,
43400 Serdang, Selangor, Malaysia Malaysia
3
Institute of Tropical Forestry and Forest Products, Universiti
Putra Malaysia, 43400 Serdang, Selangor, Malaysia

123
920 A. S. Ahmad Hazmi et al.

Graphical Abstract

Medium functionality polyol

HO OH

HO
OH
HO

Diisocyanate
Dibutyltin dilaurate

Crosslinking
RU RU RA
RA
RU
RA
RU
RA
RU RA

Polyurethane Crosslinked network with allophanate group

Keywords Polyurethane elastomer  Biomaterial 

Thermal properties  Degradation kinetic  Mechanical (a) R1-NCO + OH-R2 → R1-NHCOOR2 + Heat

properties  Frequency response

(b) R-NCO + H 2 O → [R-NHCOOH] → RNH 2 + CO 2 + Heat

(c) R1-NCO + R2-NH 2 → R1-NHCONH-R2

Introduction

(d) R1-NCO + R2COOH → [R1NHCOOCOR2] → R1NHCOR2 + CO 2

Polyurethane represents a group of versatile polymer and
conceived wide range of technical applications. Poly-
urethane ranks fifth in high volume of plastic production R1 COOR2
after polyethylene, polyvinyl chloride, polypropylene, and 1
R NHCOOR 2
+ 1
R -NCO N
polystyrene [1]. Recent impetuses for steady growth of (e)
polyurethane market are emerging applications in CONHR1

biomedical field, exceptional performance in automotive

industry, and environmental sustainability. Early genera-
tion of polyol is strongly connected to the renewable R1 CONHR1
resources [2, 3]. Current generation of polyurethane is
1
R NHCONHR 1
+ 2
R -NCO N
(f)
prepared from petroleum-based polyol which associated
with environmental and sustainability concerns. Raw CONHR2

materials derived from renewable resources are regarded as
environmental friendly and sustainable. Vegetable oil, in
particular, is a precious starting materials toward valuable
products such as high-performance polyurethane besides
being relatively inexpensive [4, 5].
Versatility of polyurethane chemistry (Fig. 1) allow
balance of hydrophilicity and hydrophobicity on the sur-
face. Polyurethanes elastomers are commonly found inside
medical microdevices, scaffolding, and biomedical micro-
electromechanical system (bioMEMS) as being
Fig. 1 Reactions between isocyanates and different compounds [2].
The individual schemes are labeled: a exothermic reaction with
alcohol which contain hydroxyl group to form urethane, b exothermic
reaction between isocyanate and water leads to unstable carbamic
acid that produce amine and gaseous carbon dioxide which is
convenient for cellular formation, c amine group reacts rapidly with
isocyanate to form disubstituted urea, d isocyanate react with
carboxylic acid to yield unstable anhydride which stabilize as amide
and release carbon dioxide, e above 80 C and under vacuum, a
reversible reaction with urethane group leads to allophanate forma-
tion, and f above 80 C and under vacuum, a reversible reaction with
disubstituted urea to form biuret group

123
Understanding intrinsic plasticizer in vegetable oil-based polyurethane elastomer as…
Fig. 2 Chemical structures of O
vegetable oil-based elastomer
and PCL. Upon polymerization,
vegetable oil-based elastomer O
will have side chains that act as
intrinsic plasticizer O
O
O
O CH3
Vegetable oil-based elastomer
biocompatible and inherit relatively lower cost than inor-
ganic materials [6]. Advantages of polyurethane elastomer
include higher machinability than glass and having more
adhesive property which makes it compatible to be used
with many materials. However, it is an inconvenient fact
that the petroleum-based elastomer; for instance, poly-
caprolactone (PCL) typically requires a substantial amount
of external plasticizer to acquire good mechanical flexi-
bility, augment crystallinity, and provide some elastomeric
mechanical properties [7–11]. Alternatively long chains of
fatty acid found in vegetable oil could be manipulated to
produce side chains which then act as intrinsic plasticizer
(Fig. 2). Furthermore, it was found that hydrophilicity, an
important key toward biocompatibility and minimizing
biofouling, of vegetable oil-based polyurethane could be
improved by dictating the hydroxyl number [12]. Hence
there are growing interests on understanding the structural
properties relationship [13–19] in renewable polyurethane.
Often decent properties for polyurethane are gained when
employing medium functionality, between 4.5 to 5.5 OH
groups/mol, as it capable to give medium conversion at gel
point and pristine crosslink densities [2]. Nevertheless,
fabricators pay little attention to the effect of crosslink
densities which eventually govern the physical properties
of elastomer.
One of promising yet underutilized among vegetable oil
for commercialization is jatropha oil. Very high and
921
CH3
OOCNH R
CH3
OOCNH R Side
chain
R HNCOO O CH3
R HNCOO OOCNH R
O
n
Polycaprolactone (PCL)
uniform contain of unsaturated fatty acids, specifically as
oleic acid and linoleic acid that constituted more than 70%
in crude oil [20], is a characteristic of jatropha oil and vital
toward achieving desired functionality and molecular
weight between crosslink (Mc). Jatropha oil is known for its
inherited toxic ingredients and naturally inedible and hence
its price relatively insulated from supply-and-demand
surge in food industry [21]. As vegetable oil-based polyol
and PCL share similarity in chemical structure that form
the repeating unit in polymeric network (Fig. 1), it is
paramount to understand the physical properties of such
vegetable oil-based polyurethane. In this paper, veg-
etable oil-based polyurethanes were prepared by reacting
multifunctional jatropha oil-based polyol with two types of
aromatic diisocyanates namely dipheylmethane-4,4’-diiso-
cyanate (MDI) and toluene-2,4-diisocyanate (TDI). MDI is
highly sought in biomedical applications due it
biodegradable characteristic. On the other hand, TDI usu-
ally render superior flexibility toward mechanical load. The
physical properties of these elastomers were examined on
differential scanning calorimeter (DSC), thermogravimet-
ric analysis (TG), and derivative thermogravimetric
(DTG), dynamic mechanical analysis (DMA), Instron
universal tensile machine, and scanning electron micro-
scopy (SEM). The crosslink densities were varied by reg-
ulating ratio of hydroxyl to isocyanates as reflected in the
swelling measurement in toluene and extend of reaction
123
922
was monitored by Fourier transform infrared (FTIR)
method.
Experimental
Materials
The multifunctional jatropha oil-based polyol used was
having hydroxyl number 175 KOH g-1 and functionality 5.1,
prepared from crude jatropha curcas oil having iodine value
104.4 g I2 (100 g-1) and functionalized via epoxidation and
ring opening [22]. Technical-grade toluene-2,4-diisocyanate
(TDI) and dibutyltin dilaurate were supplied by Fluka and
used as received. Synthesis grade dipheylmethane-4,40 -di-
isocyanate (MDI) was obtained from Merck and used as
received. Silicone-free defoamer Afcona 2020 was acquired
from AFCONA Additives and used as received.
Synthesis of jatropha oil-based polyurethane
elastomer
Jatropha oil-based polyurethane elastomer was prepared by
reacting jatropha oil-based polyol with MDI and TDI. The
molar ratios of hydroxyl group (OH) to the isocyanate
group (NCO) were 1.0/1.0, 1.0/1.1, and 1.0/1.2 as defined
in following equation:
OH Wpolyol =EWpolyol
r¼ ¼
ð1Þ
NCO WPU  Wpolyol =EWisocyanate
where Wpolyol is the mass of polyol, EWpolyol is the
equivalent mass of polyol, WPU is the mass of produced
polyurethane, and EWisocyanate is the equivalent mass of
isocyanate. The equivalent mass as supplied for TDI and
MDI are 87 g mol-1 and 125 g mol-1, respectively. A
calculated amount of diisocyanate and catalyst dibutyltin
dilaurate at 0.5 mass% were added into jatropha oil-based
polyol in a Teflon container 10 cm 9 10 cm 9 1.5 cm. As
this paper is intended to produce polyurethane elastomer,
residual water and formic acid might be a concern upon
contact with isocyanate, and hence residual gas evacuation
from casting was facilitated by the additive defoamer at
1 mass%. The mixture was stirred for 1 min and degassed
at 85 C for 30 min. The produced film was immediately
post-cured at room temperature for 24 h. Six different
formulations were used when producing jatropha oil-based
polyurethane to observe the effect of crosslink densities.
Swelling in solvent
Swelling studies on jatropha oil-based polyurethane was
determined by using toluene as solvent. Samples mass
123
A. S. Ahmad Hazmi et al.
between 0.2 and 0.6 g was immersed in 50 mL of toluene
for 10 days at 25 C. The samples were dried prior to mass
measurement. Mc was calculated from Flory–Rehner rela-
tionship [23]:
 
1=3
q2 V1 /2  2/f 2
Mc ¼ ð2Þ
lnð1  /2 Þ þ /2 þ v12 /22
where q2 is the density of polymer and f is the functionality
of network. V1 is the molar volume of solvent and in case
of toluene equal to 106.28 cm3 mol-1. /2 is the volume
fraction of polymer and calculated as follows:
m1 d1
/2 ¼ ð3Þ
m 1 ð d1  d2 Þ þ m 2 d2
m1 is the polymer mass before immersion, m2 is the
polymer mass after swollen in solvent and reach equilib-
rium, d1 is the density of solvent, and d2 is the density of
polymer. Meanwhile, polymer–solvent interaction param-
eter v12 [24] is determined from following equation:
V1 ðd1  d2 Þ2
v12 ffi b þ ð4Þ
RT
where b is lattice constant and approximated as 0.34. d1 is
the solubility parameter of solvent while d2 is solubility
parameter of polymer. R and T are universal gas constant
and temperature, respectively. As the jatropha oil-based
polyol used was ring-opened with majority methanol, v
was deliberated as 0.5293 throughout the experiment [25].
Fourier transform infrared spectroscopy (FTIR)
measurement
The FTIR spectra were recorded on Perkin-Elmer Spec-
trum 2000 spectrometer which collects mid-infrared scat-
tered radiations using a single-beam improved Michelson
interferometer with attenuated total reflectance (ATR)
accessory which equipped with diamond crystal tip. The
spectra were recorded in transmission mode in the range of
4000–500 cm-1 with a nominal resolution of 4 cm-1.
Differential scanning calorimetry (DSC)
measurement
Differential scanning calorimetry (DSC) was carried out on
DSC 823e (Mettler-Toledo), equipped with STARe soft-
ware 9.10, according to ASTM E3418-03 standard practice
[26]. The DSC equipped with a refrigerated cooling system
and calibrated with indium. Five milligrams of sample was
sealed in aluminum 40 lL hermetic pan and heated from 25
to 180 C at heating rate of 10 C min-1 to eliminate
thermal history, cooled to -60 C at cooling rate of
Understanding intrinsic plasticizer in vegetable oil-based polyurethane elastomer as… 923

5 C min-1, and heated again to 180 C min-1 at heating Results and discussion
rate of 10 C min-1 under nitrogen gas flow rate of
20 mL min-1. Physical properties of polyurethane elastomer

ATR-FTIR of polyurethane
Thermogravimetric analysis (TG/DTG)
measurement
In polymerization process, the isocyanate, which is a strong
infrared absorbent, was monitored for degree of polymer-
Thermogravimetric analysis (TG) and derivative thermo-
ization and quality of produced polyurethane [30]. The
gravimetric (DTG) were carried out on TGA/SDTA 851E
absorption intensity of the isocyanate group at 2273 cm-1
(Mettler-Toledo) according to ASTM E1641-99 standard
and hydroxyl group at 3475 cm-1 gradually decreased
practice [27]. Five milligrams of sample was put on alu-
throughout reaction time and disappeared when reaction
mina sample holder 70 lL and heated from 30 to 800 C at
time reached 5 h as shown in Fig. 3a. At the end of reac-
heating rate of 2.5, 5, 10, 15, and 20 C min-1 under
tion, four characteristic absorptions of polyurethane were
nitrogen gas flow (20 mL min-1). Small sample size
observed as NH group stretching at 3340 cm-1, urethane
equilibrated faster with furnace temperature and hence
carbonyl group C=O at 1740 cm-1, C–O stretch at
mitigated temperature deviation and noise in the mea-
1225 cm-1 and NH group bending at 1538 cm-1. It is well
surement. The TG/DTG curves were analyzed on STARe
documented that excess isocyanate and application of
software version 9.10 Mettler.
catalyst dibutyltin dilaurate would lead to formation of
allophanate and biuret [31–33]. Introduction of excess
Dynamic mechanical analysis (DMA) measurement isocyanate would enhance their formations over the ure-
thane bonds and thermodynamically favorable [34, 35].
Dynamic mechanical analysis (DMA) was carried out on This was evident from differences in wet experimental
DMA Q800 V20.24 (TA Instruments) according to ASTM
D5062-01 standard practice [28]. Instrument calibrations
were verified and a series of clamp calibrations were car- (a)
5 hour NH C = ON-H COO,
ried out on each day of measurement. A rectangular test
Transmittance/Arbitrary unit

C-O
specimen (76 by 13 by 1 mm) was clamped in tension 3 hour
mode configuration. In temperature sweep of controlled
strain program, specimens were set to -100 C using ‘‘Go 2 hour
930 cm–1
1454 cm–1
To Temperature’’ command, re-torqued, and ramped to NCO
0 hour OH
140 C at 5 C min-1 at 1 Hz according to standard
practice ASTM D5026-01. Small size specimens used in
tension mode minimize temperature gradient and allowed 4000 3500 3000 2500 2000 1500 1000 500
higher temperature scanning rate. Wavenumber/cm–1
(b)
TDI
r = 1/1.2
Tensile strength measurement
TDI
r = 1/1.1
Trasmittance/Arbitrary unit

Tensile strength of sample was analyzed using Instron 5567 TDI

universal testing machine, equipped with Bluehill version r = 1/1.0
2.14 software. For each sample, five identical dumbbell- MDI
r = 1/1.2
shaped samples were prepared by using ASTM D638-03
Type V cutter [29]. The crosshead speed was 1 mm min-1 MDI
r = 1/1.1
with 30 kN load cell. The measurement was carried out at
MDI
room temperature and 50% relative humidity. r = 1/1.0
AF2 AB2 AF1 AB1
1800 1750 1700 1650 1600 1550
Scanning electron microscope (SEM) Wavenumber/cm–1

Scanning electron microscopy were carried out using
Hitachi S-3400 N at 5 kV with 3009 magnification with
all specimens coated with gold to eliminate charging
effect.
Fig. 3 ATR-FTIR spectra of jatropha oil-based polyurethane. The
individual figures are labeled: a FTIR spectra of jatropha oil-based
polyurethane (TDI-based, r = 1/1.0) formation as a function of
reaction time, b FTIR from 1800 to 1550 cm-1 as a function of
diisocyanate and r

123
924 A. S. Ahmad Hazmi et al.

Table 1 Effect of isocyanate ratio in jatropha oil-based polyurethane elastomers toward crosslink densities, glass transition temperature, and
mechanical properties
Polyurethane Molecular Conversion at gel Glass transition temperature/ C Young modulus/ Tensile stress at Elongation at
weight between point based on MPa break/MPa break/%
crosslink/ Macosko–Miller DSC DMA
g mol-1 onset E00 peak

MDI 842 0.42 -46.9 -50.0 18.9 ± 0.7 4.1 ± 0.2 130.9 ± 5.9
r = 1/1.0
MDI 729 0.44 -45.8 -48.0 25 ± 1 4.9 ± 0.2 120.9 ± 3.8
r = 1/1.1
MDI 712 0.46 -45.7 -45.3 37.8 ± 1 5.7 ± 0.2 109.9 ± 2.2
r = 1/1.2
TDI 1017 0.50 -49.0 -55.4 11.5 ± 0.4 2.6 ± 0.1 193.8 ± 8.5
r = 1/1.0
TDI 859 0.52 -48.3 -53.1 13.9 ± 0.6 3 ± 0.2 172.8 ± 6.4
r = 1/1.1
TDI 834 0.54 -48.2 -51.2 17.8 ± 0.7 3.4 ± 0.2 162.3 ± 7.2
r = 1/1.2

analysis of crosslink densities represented by Mc values on AB1 þ AB2

RC¼O ¼ ð5Þ
Table 1 and discussed detailed in the following sec- AF1 þ AF2
tion. Typically the formation of these groups is verified
where AB1 is the surface area of bonded hydrogen bond of
using spectroscopy methods. One way is to deconvulate
carbonyl group of urea (1680–1640 cm-1), AB2 is the
FTIR data into individual contribution. The carbonyl bands
surface area of bonded hydrogen bond of carbonyl group of
of allophanate and biuret groups is known to be between
urethane (1727–1705 cm-1), AF1 is the surface area of
1460 and 1450 cm-1, and overlapped with CH2 bending of
unbonded hydrogen bond of carbonyl group of urea (1701–
polyol. To differentiate the bands, we measured the
1690 cm-1), and AF2 is the surface area of unbonded
absorption ratio between 1454 and 930 cm-1 where
hydrogen bond of carbonyl group of urethane (1745–
930 cm-1 was attributed to ether group [36]. Formation of
1736 cm-1). Then the values of DPS can be calculated as:
branching allophanate and biuret groups in the produced
polyurethane elastomers were verified via excess FTIR RC¼O
DPS ¼ ð6Þ
absorption ratio between 1454 and 930 cm-1, i.e., above 1 þ RC¼O
1.4. Ester and ether formations were observed as strong
An appreciable amount of DPS was found in all pre-
absorption peaks at 1100–1200 cm-1. Ester group was
pared elastomers (Table 2). Phase separation phenomena
located along the fatty acid chains, urethane, allophanate,
and interpenetration in multi-component polymer are
and biuret groups. On the other hand ether group was
attributed to incompatibility between the soft and hard
produced in vicinity of side chain. This ether group
segments. DPS tended to be disproportionate with
specifically was also known as methoxy group and origi-
increasing crosslink densities. The concentration of
nated from methanol during ring-opening step in producing
hydrogen bonds in elastomers with higher isocyanate index
polyol. Besides, the casting mixtures were solvent free and
(i.e., MDI (r = 1/1.2) and TDI (r = 1/1.2)) was found to
produced transparent polyurethane sheets that reflected the
be as much as 30% higher in comparison to elastomers
color of the polyol.
with lower isocyanate index (i.e., MDI (r = 1/1.0) and TDI
To further investigate phase existed in the elastomers,
(r = 1/1.0)). TDI-based elastomer shown little increment
one need to determine the RC=O index and subsequent
due having fewer free hydrogen in single aromatic struc-
degree of phase separation (DPS). Phase separation
ture whereas MDI had more free hydrogen in its double
describes the sharing of hard segment with bounded
aromatic structures.
hydrogen bonds. By using FTIR data from each sample, the
carbonyl stretching zone between 1770 and 1640 cm-1
Crosslink densities
was corrected by subtracting the baseline. Then the mul-
tiplet bands were deconvulated (Fig. 3b) using Guassian
A crosslinked polymer is able to increase its volume by
curve-fitting into individual band by using Origin 8.5.0.
absorbing large amount of solvent without being dissolved
This allowed one to calculate RC=O index as [36, 37]:

123
Understanding intrinsic plasticizer in vegetable oil-based polyurethane elastomer as… 925

Table 2 Deconvolution results of FTIR of jatropha oil-based polyurethane elastomers between 1745 to 1650 cm-1, and subsequent degree of
phase separation (DPS)
Polyurethane AF2 AB2 AF1 AB1 Carbonyl hydrogen Degree of phase
1745–1736 cm-1 1727–1705 cm-1 1701–1690 cm-1 1680–1650 cm-1 bonding index, RC=O separation, DPS/%

MDI 0.06294 0.02641 0.02458 0.12137 1.6885 62.8

r = 1/1.0
MDI 0.10824 0.03863 0.05817 0.15703 1.1758 54.0
r = 1/1.1
MDI 0.07487 0.02139 0.05223 0.09751 0.9355 48.3
r = 1/1.2
TDI 0.07187 0.01716 0.04304 0.13791 1.3494 57.4
r = 1/1.0
TDI 0.09836 0.00186 0.07049 0.21273 1.2709 56.0
r = 1/1.1
TDI 0.08213 0.07575 0.06309 0.10832 1.2675 55.9
r = 1/1.2

as the crosslink sites comprehend the increased inter-
molecular strands. Crosslink density is defined as 1/(2 Mc)
where Mc is the number average molecular weight between
crosslink. A crosslinked network has nearly infinite
molecular weight, and hence Mc value is more specific in
denoting network properties. A minimal value of Mc
indicates tighter packing density with fewer polymers per
crosslink. As shown in Table 1, jatropha oil-based poly-
urethane elastomer at higher isocyanate ratio showed lower
Mc and hence higher crosslink densities. This is also
reported by Dey et al. [38] where lower Mc was achieved
when more isocyanate incorporated into polymeric net-
work. MDI-based polyurethane had higher crosslink den-
sities than TDI-based polyurethane due to the symmetrical
structure and allowed tighter packing organization. Due to
the nature of high reactivity of aromatic diisocyanate [31],
excess isocyanates may react with urethane and urea
groups to afford branching allophanate and biuret groups,
which increased overall crosslink densities [39]. Based on
Macosko–Miller polyfunctional equation [40], conversions
in the range from 0.42 to 0.54 as obtained in these exper-
iments (Table 1) were close to midpoint and typically yield
polymer with enhanced physical properties [2].
Glass transition temperature (Tg) by DSC and DMA
The glass transition temperature (Tg) analysis was done
using DSC (Fig. 4) and DMA (Fig. 5). Values of Tg
observed in both DSC and DMA were comparable and in
the range from -55 to -45 C (Table 1). Tg obtained in
DSC and DMA were the a-relaxation where phase transi-
tion occurred in bulk. It is anticipated that the b-relaxation
occurred further down from the DMA measurement range,
which was below -100 C. The onset of DSC curves
represented Tg and increased with increasing crosslink
densities. The successive small exothermic peak appeared
between -16 and 2 C reflects small amount of rear-
rangement within network structure and probably origi-
nated from the rotational motion of fatty acid [41].
Meanwhile, in the temperature sweep of DMA, the storage
modulus (E0 ) decreased with decreasing crosslink density
(Fig. 5a). The loss modulus (E00 ) shows peak in the range
from -55 to -45 C, and the peak height decrease with
decreasing crosslink density (Fig. 5b). Broader area under
the E00 peak suggests a softer material and wider molecular
weight distribution. The TDI-based exhibited longer phase
lag than MDI-based attributed by less-ordered packing
structure. The difference between Tg values obtained from
DSC and DMA as shown in Table 1 originated from the
choice of frequency employed in DMA measurement. Note
that in the measurement range there was no apparent
melting point observed in the DSC curves which suggested
the samples were thermoset. The majority amorphous
domain probably explained why the samples were
transparent.
There are two postulates to the subambient Tg obtained.
First is the application of high molecular weight of multi-
functional polyol. Introduction of long-chain polyol
reduced the concentration of hard segments like urethane
and urea bonds, thus lowering the cohesive interaction
between these hard segments and subsequently lowered the
Tg of network formed. Second as a consequence for sig-
nificant amount of side chains (Fig. 2) that reduced inter-
chain interaction, minimized crystallization of soft seg-
ments, and consequently lowered the tensile strength and
modulus. Furthermore MDI-based polyurethane showed
higher Tg than TDI-based. The latter exhibited a larger
range of Tg and suggests a wider distribution of crosslink

123
926 A. S. Ahmad Hazmi et al.

(a) (a)
0.5 8000

7000 MDI (r = 1/1.0)

Storage modulus/MPa
Heat flow/mW Endo↔Exo

0 MDI (r = 1/1.0) MDI (r = 1/1.1)

6000
MDI (r = 1/1.1) MDI (r = 1/1.2)
–0.5 5000
TDI (r = 1/1.0)
MDI (r = 1/1.2) 4000 TDI (r = 1/1.1)
–1 TDI (r = 1/1.2)
3000

–1.5 2000

1000
–2 0
–100 –50 0 50 100 150
–2.5 Temperature/°C
–60 –10 40 90 140 190
(b)
Temperature/°C 400
(b) 350
2 MDI (r = 1/1.0)

Loss modulus/MPa
300
MDI (r = 1/1.1)
Heat flow/mW Endo↔Exo

1
250 MDI (r = 1/1.2)
0 TDI (r = 1/1.0)
TDI (r = 1/1.0)
TDI (r = 1/1.1) 200
TDI (r = 1/1.1)
–1 TDI (r = 1/1.2) 150 TDI (r = 1/1.2)
–2 100
–3 50

–4 0
–100 –50 0 50 100 150
–5 Temperature/°C

–6
–60 –10 40 90
Temperature/°C
Fig. 4 DSC curve of jatropha oil-based polyurethane elastomers at
10 C min-1 in nitrogen atmosphere. The individual curves are
labeled a MDI-based polyurethane, b TDI-based polyurethane
densities. The symmetrical structure MDI increased the
tendency of polyurethane to crystallize and had better
packing of rigid segment, hence the enhanced tensile
strength and modulus. Fox–Loshaek equation anticipates
linear relationship between glass transition of crosslinked
polymer (Tg), uncrosslinked polymer (Tgo), a constant (k),
and crosslink density as follow [13, 42]:
k groups to afford branching allophanate and biuret groups
Tg ¼ Tgo þ
Mc
An increase in the molar ratio of isocyanate to hydroxyl
had induced the increment of Tg values as isocyanate
groups were the main constituents of rigid segments. For a
lightly crosslinked polymer such as the elastomers pro-
duced in this study, it was observed that there was a linear
increase of Tg with increasing crosslink density (Fig. 6).
The slopes for MDI- and TDI-based polyurethane elas-
tomers were 18.2 and 17.1 C mol kg-1, respectively. In
other word, the Tg of MDI-based polyurethane elastomers
showed more dependent on crosslink density than TDI-
based polyurethane elastomers. This higher sensitivity
inherited by MDI-based polyurethane is probably owed to
symmetrical structure of MDI that allow better molecular
140 190
Fig. 5 DMA response of jatropha oil-based polyurethane from -100
to 140 C at 5 C min-1 and 1 Hz. The individual curves are labeled
a the storage modulus (E0 ) of jatropha oil-based polyurethane as a
function of temperature at 1 Hz, b the loss modulus (E00 ) of jatropha
oil-based polyurethane as a function of temperature at 1 Hz
alignment and higher degree of crystallization as value of
Mc is known sensitive to network imperfection. Further
higher isocyanate ratio led to higher concentration of
branching allophanate and biuret. This in turn produced
elastomer with higher mechanical modulus and improved
thermal stability. Due to the nature of high reactivity of
aromatic diisocyanate [31] and high temperature reaction,
excess isocyanates may reacted with urethane and urea
ð7Þ (Fig. 1e, f), which increased overall crosslink densities
(Table 1). According to Eq. (7), sufficiently very large
value of Mc would lead decrease value of Tg to approach
Tgo. From Fig. 6, both MDI and TDI-based polyurethanes
have Tgo about -72 C. One way to approach this lower
value of Tg is by extending the side chain of polyol as
demonstrated by Clark AJ and Hoong SS [43]. Note that
they employed a high mass percent of tetrahydrofuran
(THF) incorporation in the side chain and had obtained Tg
as low as -76 C.
The effect of these supplementary branching groups
probably can be explained from difference found in qual-
itative approach of frequency response in DMA where E0
and complex viscosity (g*) change as a function of fre-
quency (Fig. 7) based on Doi-Edwards theory [44]. It is

123
Understanding intrinsic plasticizer in vegetable oil-based polyurethane elastomer as… 927

–44 –50 1000 100

MDI-based TDI-based MDI -based (r = 1/1.2)

10
–46

Log η*/MPa S–1

–52 100

Log E' /MPa

Tg /°C

Tg /°C
1

–48
10
–54
0.1
E' of MDI (r = 1/1.2) TDI -based (r = 1/1.2)
E' of TDI (r = 1/1.2)
–50
n* of MDI (r = 1/1.2)
n* of TDI (r = 1/1.2)
1 0.01
–56 0.1 1 10 100
1.2 1.4 1.0 1.2
Log frequency, Hz
1/MC /mol kg–1 1/MC /mol kg–1
Fig. 7 Crossovers between storage modulus (E0 ) and complex
Fig. 6 A linear dependence between glass transitions temperature viscosity (g*) for MDI-based and TDI-based elastomers at r =
(Tg) and crosslink density 1/1.2 which correspond to relative molecular weight and molecular
weight distribution

known that the difference found in frequency response

7
arise from the network structure. At very low frequency, MDI
TDI
r = 1/1.2
MDI-based samples had higher g* than TDI-based which 6 r = 1/1.2
TDI
suggests more branching across the polymeric chain. The MDI
r = 1/1.1
5 r = 1/1.1
crossover between the E0 - g* for MDI-based samples was TDI
Tensile stress/MPa

MDI r = 1/1.0
higher than TDI-based, indicating qualitatively higher 4 r = 1/1.0
molecular weight. Meanwhile, the crossover between the
3
E0 - g* of TDI-based occur at much higher frequencies
which suggest broader molecular weight distribution. 2
Long-branched polymeric chains such as MDI-based
1
(r = 1/1.2) and TDI-based (r = 1/1.2) showed more fre-
quency dependent and reflected in the E0 curve as reported 0
by Kevin [45]. In retroperspective, higher E0 in MDI-based 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Tensile strain/mm mm–1
also reflect higher ability to store energy.
Fig. 8 Effect of crosslink densities on mechanical tensile stress–
Tensile stress–strain behavior of polyurethane elastomer strain curves for jatropha oil-based polyurethane elastomers at 25 C

As the Tg for all jatropha oil-based polyurethane well
below the room temperature, they exhibited highly elastic
mechanical properties (Fig. 8). Polyurethane elastomers
derived from MDI showed higher tensile strength than TDI
(Table 1). The ultimate elongation before break for speci-
mens in Table 1 was up to 131 and 194% for MDI-based
and TDI-based, respectively. Very low Tg and decent
mechanical modulus are sought in producing scaffolding
intended for biomedical engineering [46] or in engineering
sectors to avoid unanticipated failure. Jatropha oil-based
polyurethane showed nonlinear brittle-like tensile stress–
strain curve with greater flexibility. It is worthy to note that
the produced elastomers had exceeded 100% tensile elon-
gation despite behaving as brittle-like in the tensile stress–
strain. It was observed that, under tensile strain, there was
little visible necking which showed the complexity of
mechanical integrity. It was thought that the available
energy storage, contributed by the crosslinked polymeric
network and entrapped in the chain entanglement, was able
to compensate physical yield in the load direction without
appreciable deformation in transverse direction. Other type
of vegetable oil such as palm oil had also demonstrated
similar mechanical capability as elastomer [43]. The stiff
initial response at the beginning of loading is a character-
istic of jatropha oil-based polyurethane, corresponding to
the cleavage of the hydrogen bonds and other secondary
forces in the soft segments [33]. Further the application of
aromatic diisocyanate had enhanced the tensile strength.
On the other hand, one can increase the elongation at break
by employing aliphatic diisocyanate [47]. This, however,
may depreciate mechanical modulus of the elastomers.
In an unstrained state, the polymeric chains are highly
coiled with random dispersion. As the polyurethane
underwent strain, the soft segment disentangled and

123
928 A. S. Ahmad Hazmi et al.

straightened. Additional elongation diminishes the intrinsic

cohesive energy and structures of polyol and rigid seg-
ments reoriented normal to the applied stress. The ori-
entability depends on the isocyanate ratio as well as the
symmetry of isocyanate and confers capability to oppose
further elongation. Application of long chain in the soft
segment had afforded extensive plastic deformation. One
remarkable feature of the produced polyurethane in this
paper, which basis of chemical structure akin to PCL
(Fig. 2), was able to show highly elastic mechanical
deformation. Typically PCL content in polyurethane
deterred this type of mechanical response. Polyurethane
elastomers in this study had somewhat opposite mechanical
behavior with PCL-derived elastomers [7, 48, 49], albeit
having comparable Tg and tensile strength. Fig. 9 Scanning electron fractrograph showing brittle fracture result-
ing from uniaxial tensile load of jatropha oil-based polyurethane
Polyurethanes network with higher crosslink densities elastomer (magnification 3009, 5 kV). The individual fractographs
had increased mechanical Young modulus and strength at are labeled a MDI-based with r = 1/1.0, b MDI-based with r =
the expense of elongation. MDI-based elastomers, which 1/1.1, c MDI-based with r = 1/1.2, d TDI-based with r = 1/1.0,
had higher Tg and lower values of DPS, exhibited e TDI-based with r = 1/1.1, f TDI-based with r = 1/1.2. The arrows
indicate the ridges which propagated from the origins of the crack
approximately twofold increment in tensile modulus and with radial fan-shaped patterns corresponded to change in orientation
strength and shorter elongation at break. On the other hand, of the cleavage of atomic bonds at specific planes
TDI-based elastomers had higher values of DPS, suggest-
ing less hard domain crystallinity and attained outstanding
elongation at break. The crosslink sites are analogous to Table 3 Onset degradation temperature and DTG temperature at
maximum decomposition rate for jatropha oil-based polyurethane
vulcanization and act as strong anchor points to restrain elastomers
neighboring polymeric chains from slippage during defor-
Polyurethane MDI MDI MDI TDI TDI TDI
mation and depend primarily on the polyol characteristic
r = 1/ r = 1/ r = 1/ r = 1/ r = 1/ r = 1/
[2]. The majority soft segment comprised ether group 1.0 1.1 1.2 1.0 1.1 1.2
where C–O bond is easier to rotate and confer the molec-
ular flexibility. On the other hand, aromatic rings in the Onset of thermal degradation/C
hard segment limit molecular flexibility due to crystal- T2% 200 222 249 190 211 222
lization and steric hindrance. The existence of side chains T5% 264 286 291 248 254 275
which typically associated with impaired inter-molecular T10% 291 312 323 275 280 296
crystallization was found not to be detrimental to the ten- DTG temperature at maximum decomposition rate/C
sile strength. The fractographic examination (Fig. 9) using Step 1 293 301 303 281 287 286
scanning electron microscopy (SEM) revealed that the Step 2 374 383 399 392 388 391
fracture surfaces were relatively rough and irregular which Step 3 465 471 476 – – –
typically associated with brittle fracture. Appreciable gross
deformation such as twisting and tearing became apparent
for higher crosslink densities elastomers (Fig. 9c, f). MDI- based polyurethanes showed higher onset degradation and
based elastomers showed lower degree of tearing and rid- more receptive with isocyanate ratio than TDI-based
ges due to higher rigidity of molecular arrangement, while (Fig. 10). All samples showed increasing onset degradation
TDI-based elastomers showed more tearing due to greater temperature with increasing crosslink densities (Table 3).
plastic deformation and appeared clearly with increasing MDI-based elastomer had onset degradation temperature at
isocyanate ratio. 2% mass loss (T2%) remarkably increased by 49 C for
higher crosslink densities MDI-based (r = 1/1.2) against
Thermal properties of polyurethane elastomer lower crosslink densities MDI-based (r = 1/1.0). This was
due to less urethane concentration in the higher crosslink
TG and DTG analysis densities samples. MDI-based samples exhibited higher
onset temperature than TDI-based due to better contribution
The produced elastomer demonstrated relatively high ther- of crosslink densities. On the other hand, TDI-based samples
mal stability with onset degradation temperature at 5% mass exhibited less dependent on isocyanate ratio as illustrated by
loss (T5%) ranging from 248 to 291 C (Table 3). MDI- a small shift in maximum decomposition temperatures.

123
Understanding intrinsic plasticizer in vegetable oil-based polyurethane elastomer as… 929

Fig. 10 TG curves of jatropha 200 0.0000

oil-based polyurethane at constant 0.0000

heating rate of 10 C/min under

180
nitrogen atmosphere. DTG is the –0.0005 –0.0005
derivative of TG curve, indicating
thermal decomposition rate. Inset 160

DTG/s–1
–0.0010 3
is the deconvulation DTG curve of –0.0010
TDI-based polyurethane 140 –0.0015 2
elastomer. The individual numbers 1
of inset are labeled: 1 multiplet –0.0015

DTG/s–1
120 –0.0020

band, 2 cumulative fit peak of 200 300 400 500

Mass/%
Temperature/°C
Gaussian, and 3 individual
Gaussian peaks 100 –0.0020

80
–0.0025

60 MDI (r = 1/1.0)
MDI (r = 1/1.1) –0.0030
40 MDI (r = 1/1.2)
TDI (r = 1/1.0)
TDI (r = 1/1.1) –0.0035
20 TDI (r = 1/1.2)

0 –0.0040
0 200 400 600
Temperature/°C

Application of long-chain polyol derived from vegetable oil D

relinquished inferior thermal stability typically associated
with soft segment. Thermal dissociation of allophanate
reproduced urethane and isocyanate while dissociation of
biuret regenerated disubstituted urea and isocyanate. It was
suggested that the branching allophanate and biuret formed
in this study was capable to resist initial thermal breakdown.
In DTG curves (Fig. 10), MDI-based samples had three
degradation steps while TDI-based had two which agreed
to other findings [36, 50, 51]. Table 3 shows the temper-
ature at maximum decomposition rate as obtained in DTG
curves. In the first step, low rate of the mass loss was
associated with depolymerization including dissociation of
branching allophanate and biuret groups. The second and
third step involved higher rate of the mass loss with evo-
lution of volatile compounds such as carbon dioxide, car-
bon monoxide, and nitriles [52]. Initial mass loss is usually
began with the decomposition of urethane bond into carbon
dioxide and isocyanate components (Eqs. 9–11) [51]. It is
known that the main thermal breakdown mechanism of
urethane is by scission of polyol-isocyanate bond. At
higher temperature, formations of primary and secondary
amine occur and associated with isocyanurate rings and
carbodiimide linkage [33, 53]. Typically the decomposition
of urethane is described as following reactions [33]:
D
R0 NHCOOR00 ! R0 NCO þ R00 OH
D was probably owed to functionality and heterogeneity in
R0 NHCOOCR002 CH2 R000 ! R000 CH
¼ CR002 þ R0 NH2 þ CO2
R0 NHCOOR ! R0 NHR00 þ CO2 ð10Þ
where R0 is aromatic substituent and R00 and R¢¢¢ are ali-
phatic substituent.
The DTG curves were a combination of such individual
profile of each component and could be deconvoluted
(inset of Fig. 10) to separate peak maxima by using Origin
8.5.0. The cumulative fit peak agreed well with DTG curve
with coefficient of determination (COD) above 0.99. There
are three important findings as illustrated in Table 4. First
MDI-based elastomer showed substantial improvement at
higher crosslink densities (r = 1/1.2). The peak area shif-
ted to higher temperature and MDI-based (r = 1/1.2) had
more peak area at the third step as indicated by having
highest normalized peak area. In other word MDI-based
(r = 1/1.2) had enhanced concentration of rigid segment
bounded with hydrogen bond and better molecular
arrangement as suggested by DPS (Table 2). Hence MDI-
based exhibited exceptional improvement in physical
properties. Secondly, in contrast to MDI-based, there was
no appreciable shift of peak area for TDI-based elastomer
which suggested that thermal stability of TDI-based was
not a strong function of crosslink densities. Thirdly, it is
worthy to note that deconvolution of TDI-based elastomers
prepared in this paper required three peaks to achieve
COD [ 0.99. This was not clearly presented in the original
ð8Þ DTG curves. The existence of the third peak for TDI-based
functional group positions.
ð9Þ

123
930 A. S. Ahmad Hazmi et al.

Table 4 Deconvolution results of DTG curves of jatropha oil-based polyurethane elastomers

MDI r = 1/1.0 MDI r = 1/1.1 MDI r = 1/1.2 TDI r = 1/1.0 TDI r = 1/1.1 TDI r = 1/1.2

Peak temperature/C
Step 1 290 299 314 288 293 296
Step 2 373 390 398 391 392 391
Step 3 459 469 473 453 453 452
Peak areaa/%/Normalized peak areab
Step 1 15/1 8/0.5 14/1.5 19/1.5 19/1.4 21/1.6
Step 2 40/2.8 47/3.2 29/3.1 60/4.6 63/4.6 59/4.5
Step 3 45/3.1 45/3.0 57/6.1 21/1.6 18/1.3 20/1.5
Peak width/C
Step 1 54 35 54 62 64 69
Step 2 62 74 58 54 51 49
Step 3 46 34 45 34 36 35
CODc 0.9962 0.9971 0.9976 0.9960 0.9948 0.9957
a
Peak area = Peak area*100/Total peak area
b
Normalized Area = Peak area/first peak area of MDI r = 1/1.0
c
Coefficient of determination (COD) is the measurement of goodness of fit

400 400 400

MDI MDI MDI
350 r = 1/1.0 350 r = 1/1.1 350 r = 1/1.2
Activation energy/kJ mol–1

Activation energy/kJ mol–1

Activation energy/kJ mol–1

300 300 300

250 250 250

200 200 200

150 150 150

100 100 100

0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
Conversion Conversion Conversion
300 300 300
TDI TDI TDI
r = 1/1.0 r = 1/1.1 r = 1/1.2
Activation energy/kJ mol–1
Activation energy/kJ mol–1

Activation energy/kJ mol–1

250 250 250

200 200 200

150 150 150

100 100 100

0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
Conversion Conversion Conversion

Fig. 11 Dependence of activation energy (Ea) on conversion (a) as calculated based on Ozawa method

123
Understanding intrinsic plasticizer in vegetable oil-based polyurethane elastomer as… 931

Kinetic of thermal degradation
Via isoconversional method, the temperatures correspond
to constant conversion are defined from the resultant mass
loss curves. The extent of conversion (a) at a given tem-
perature is defined as:
mi  mT
a¼ ð11Þ
mi  mf
where mi is the initial mass of step changes, mT is the mass
of isoconversion at temperature T, and mf is the final mass
of step changes. The rate equation is given by:
da
¼ kðT Þf ðaÞ
dt
where k(T) is rate constant and f(a) is the reaction model.
Temperature dependence of Arrhenius expression gives
k(T) = A exp(-Ea/RT) where A is the pre-exponential
factor, Ea is the activation energy, R is the universal gas
constant, and T is temperature at conversion. Ea, A, and
reaction order (n) are usually derived in kinetic analysis.
The experimental procedures were carried out based on
isoconversional at five different heating rates. Then the
kinetic parameters were determined by applying Ozawa
[54], Flynn–Wall, and Kissinger [55] methods. Different
values of Ea at different a verified that jatropha oil-based
polyurethane decomposed in multiple steps in Ozawa plot
(Fig. 11).
The decomposition of elastomers in this paper under the
given experimental condition can be best described by a
model assuming a restricted nucleation growth mechanism
such as n-dimensional Avrami-Erofeev solid-state model
[54] which integrated ingestion of phantom nuclei and
coalescence as summarized in Table 5. The shape of Ea-a
curve (Fig. 11) typically reflects the associated kinetic
mechanism. At the beginning was diffusion-controlled
regime with concave-upward curve and had n less than 1 for
a \ 0.1. Diffusion is a rate-limited step governed by trans-
port phenomena which included formation of product at the
solid–gas interface. It was perceived at the same a, elastomer
(r = 1/1.2) had higher reaction order indicating higher order
ð12Þ
of chemical cleavage of branching groups along the poly-
meric network. Diffusion type of thermal degradation, onset
degradation temperature, and activation energy for the first
step (a = 0.1) were characteristic of thermally unsta-
ble urethane, allophanate and biuret groups. A wide region
between 0.1 \ a \ 0.6 indicated a gradual transition from
diffusion-type to pseudo-type. This transition was evidenced
on increasing values of Ea and n. Acceleration on Ea-a curve
typically associated with thermal breakdown along weaker
bond and formation of free radical. Higher values of Ea
beyond the middle step (a [ 0.6) with concave-upward
curve were possibly contributed from relatively medium
network functionality. Nevertheless, high molecular weight
of soft segment had contributed to enhanced Ea.

Table 5 Kinetic of thermal degradation analysis based on isoconversional method for jatropha oil-based polyurethane elastomers
Polyurethane Activation energy (Ea)/kJ mol-1/pre-exponential factor (ln A)/min-1/reaction order (n)a
Flynn–Wallb Ozawab Kissingerc
a = 0.05 Step 1, a = 0.1 Step 2, a = 0.6 Step 3, a = 0.9 Step 1, a = 0.1 Step 2, a = 0.6 Step 3, a = 0.9

MDI 136/14.5/0.49 126/30.0/0.56 199/37.3/1.13 247/43.4/2.36 156/32.7/1 194/35.5/1 252/40.4/1

r = 1/1.0
MDI 170/17.6/0.44 165/38.5/0.59 302/54.6/1.27 325/55.6/2.65 164/34.6/1 220/39.8/1 292/46.7/1
r = 1/1.1
MDI 208/21.2/0.66 184/43.2/0.54 343/62.8/0.97 348/59.7/2.84 176/36.5/1 258/47.4/1 327/52.9/1
r = 1/1.2
TDI 144/15.4/0.41 145/34.4/0.69 181/35.3/0.88 261/46.4/2.42 141/29.9/1 185/32.8/1 –
r = 1/1.0
TDI 148/15.9/0.45 164/38.5/0.64 186/38.3/1.05 277/49.5/2.46 139/29.4/1 175/31.2/1 –
r = 1/1.1
TDI 154/16.3/0.75 174/40.6/0.67 201/39.4/0.89 273/49.5/2.26 124/26.4/1 152/26.7/1 –
r = 1/1.2
 
a 1 Ea

1
da


Determined from n-dimensional Avrami-Erofeev model and approximated as ln n1 ¼ R T þ ln A  ln dt
½nð1aÞ½ lnð1aÞ n
b
Isoconversional method of Ozawa and Flynn–Wall apply DoylE0 s approximation as lnðbi Þ ¼ const  RTEaa;i . The Ea is obtained as the slope in
the plot of ln qi versus 1/Ta,i as described in ASTM E1641-99
   
c Ea b n1 Ea
To determine A and n, the rate equation based on Kissinger was solved as RT 2 ¼ Anð1  aÞm exp RT m
and Ea is represented by the slope of
m

the plot of ln (b/T2m) versus 1/Tm

123
932
Furthermore, the side chains and heterogeneity in polyol
structure may introduced steric hindrance and made
decomposition processes more difficult. The last degrada-
tion step (0.7 \ a \ 0.8) was dominated by second-order
reaction and closely related to C–C bond dissociation. De-
acceleration of degradation which showed convex-upward
at final step (a [ 0.8) was related with char formation. The
values of n verified the consecutive nature of degradation
of the elastomer. The wide variation of Ea is characteristic
of complexity for continuous degradation mechanism and
summarized in Table 5. The TG/DTG curves were com-
parable to soybean, castor [51] and PCL [56].
Conclusions
A new class of enhanced bio-polyurethane elastomer was
successfully developed from medium functionality non-ed-
ible vegetable oil-based polyol which both chemical struc-
ture and physical properties comparable to PCL-derived
polyurethane. Effects of polyol network structure toward
polyurethane physical properties at different crosslink den-
sities were characterized by thermal and mechanical analy-
sis. Application of medium functionality and high molecular
weight polyol had beneficially revamped the polymer net-
work formation and physical properties. Produced elas-
tomers had highly elastic mechanical behavior and very low
Tg which was vital to conserve resilience at low temperature
or to withstand a high frequency environment. The thermal
and mechanical properties demonstrated tunable properties
by resolving the crosslink densities. The produced veg-
etable oil-based polyurethanes were thermally stable and
had relatively high onset degradation temperature.
Acknowledgements This work is supported by the Graduate
Research Fellowship, Universiti Putra Malaysia. The authors
acknowledge personnel at Universiti Putra Malaysia, Universiti
Kebangsaan Malaysia, Malaysian Nuclear Agency and Malaysian
Palm Oil Board for their expert contribution and research facilities.
The authors thank Director of Advanced Oleochemical Technology
Division and Nurul Farhana Omar for kind proofread this work.
References
1. Modern Plastic International. Mod Plast Int. Lausanne;
1990;20(1):31.
2. Ionescu M. Chemistry and technology of polyols for poly-
urethanes. Shawbury: Rapra Technology; 2005. p. 13–50.
3. Desroches M, Escouvois M, Auvergne R, Caillol S, Boutevin B.
From vegetable oils to polyurethanes: synthetic routes to polyols
and main industrial products. Polym Rev. 2012;52(1):38–79.
doi:10.1080/15583724.2011.640443.
4. Lligadas G, Ronda JC, Galià M, Cádiz V. Plant oils as platform
chemicals for polyurethane synthesis: current state-of-the-art.
Biomacromol. 2010;11(11):2825–35. doi:10.1021/bm100839x.
123
A. S. Ahmad Hazmi et al.
5. Malaysian Palm Oil Board. Malaysian Oil Palm Statistics 2012.
32 ed. Bangi; 2013.
6. Saliterman S. Fundamentals of BioMEMS and medical
microdevices. Washington: SPIE Press; 2006.
7. Loh XJ, Colin Sng KB, Li J. Synthesis and water-swelling of thermo-
responsive poly(ester urethane)s containing poly(e-caprolactone),
poly(ethylene glycol) and poly(propylene glycol). Biomaterials.
2008;29(22):3185–94. doi:10.1016/j.biomaterials.2008.04.015.
8. Woodruff MA, Hutmacher DW. The return of a forgotten poly-
mer—polycaprolactone in the 21 st century. Prog Polym Sci.
2010;35(10):1217–56. doi:10.1016/j.progpolymsci.2010.04.002.
9. Cherng JY, Hou TY, Shih MF, Talsma H, Hennink WE. Poly-
urethane-based drug delivery systems. Int J Pharm.
2013;450(1–2):145–62. doi:10.1016/j.ijpharm.2013.04.063.
10. May-Hernández L, Hernández-Sánchez F, Gomez-Ribelles JL,
Sabater-i Serra R. Segmented poly(urethane-urea) elastomers
based on polycaprolactone: structure and properties. J Appl
Polym Sci. 2011;119(4):2093–104. doi:10.1002/app.32929.
11. Nair LS, Laurencin CT. Biodegradable polymers as biomaterials.
Prog Polym Sci. 2007;32(8–9):762–98. doi:10.1016/j.progpo
lymsci.2007.05.017.
12. Saalah S, Abdullah LC, Aung MM, Salleh MZ, Awang Biak DR,
Basri M, et al. Waterborne polyurethane dispersions synthesized
from jatropha oil. Ind Crops Prod. 2015;64:194–200. doi:10.
1016/j.indcrop.2014.10.046.
13. Petrović ZS. Polyurethanes from vegetable oils. Polym Rev.
2008;48(1):109–55. doi:10.1080/15583720701834224.
14. Petrović ZS, Guo A, Javni I, Cvetković I, Hong DP. Polyurethane
networks from polyols obtained by hydroformylation of soybean
oil. Polym Int. 2008;57(2):275–81. doi:10.1002/pi.2340.
15. Petrovic ZS, Javni I, Guo A, Zhang W, inventors; Method of
making natural oil-based polyols and polyurethanes therefrom.
US Patent 6686435 patent 6686435. 2004.
16. Zlatanić A, Lava C, Zhang W, Petrović ZS. Effect of structure on
properties of polyols and polyurethanes based on different veg-
etable oils. J Polym Sci Part B Polym Phys. 2004;42(5):809–19.
doi:10.1002/polb.10737.
17. Petrović ZS, Yang L, Zlatanić A, Zhang W, Javni I. Network
structure and properties of polyurethanes from soybean oil.
J Appl Polym Sci. 2007;105(5):2717–27. doi:10.1002/app.26346.
18. De Genova R, Malsam J, Zlatanic A, Wazirzada Y. Blo-based
polyols for the flexible slabstock foam industry. J Oil Palm Res.
2008; Spec. Iss. October: 53–60.
19. Badri KH, Ahmad SH, Zakaria S. Production of a high-func-
tionality RBD palm kernel oil-based polyester polyol. J Appl
Polym Sci. 2001;81(2):384–9. doi:10.1002/app.1449.
20. Gübitz GM, Mittelbach M, Trabi M. Exploitation of the tropical
oil seed plant Jatropha curcas L. Biores Technol.
1999;67(1):73–82. doi:10.1016/s0960-8524(99)00069-3.
21. Ahmed WA, Salimon J. Phorbol ester as toxic constituents of
tropical Jatropha curcas seed oil. Eur J Sci Res. 2009;31(3):429–36.
22. Hazmi ASA, Aung MM, Abdullah LC, Salleh MZ, Mahmood
MH. Producing Jatropha oil-based polyol via epoxidation and
ring opening. Ind Crops Prod. 2013;50:563–7. doi:10.1016/j.
indcrop.2013.08.003.
23. Flory PJ, Rehner J. Statistical mechanics of cross-linked polymer
networks I. Rubberlike Elast J Chem Phys. 1943;11(11):512.
24. Krevelen DWV, Nijenhuis KT. Properties of polymers: their
correlation with chemical structure; their numerical estimation
and prediction from additive group contributions. 4th ed. Ams-
terdam: Elsevier Science; 2009.
25. Petrović ZS, Guo A, Zhang W. Structure and properties of
polyurethanes based on halogenated and nonhalogenated soy–
polyols. J Polym Sci Part A Polym Chem. 2000;38(22):4062–9.
doi:10.1002/1099-0518(20001115)38:22\4062:aid-pola60[3.0.
co;2-l.
Understanding intrinsic plasticizer in vegetable oil-based polyurethane elastomer as…
26. ASTM Standard D3418-03 I. Standard test method for transition
temperatures and enthalpies of fusion and crystallization of
polymers by differential scanning calorimetry. West Con-
shohocken, PA; 2003.
27. ASTM Standard E-1641-99 I. Standard test method for decom-
position kinetics by thermogravimetry. West Conshohocken, PA;
1999.
28. ASTM Standard D5026-01 I. Standard test method for plastics:
dynamic mechanical properties: In: Tension. West Con-
shohocken, PA; 2001.
29. ASTM Standard D638-03 I. Standard test method for tensile
properties of plastics. West Conshohocken, PA; 2003.
30. Nishikida K, Coates J. Infrared and Raman analysis of polymers.
In: Lobo H, Bonilla JV, editors. Handbook of plastics analysis.
CRC Press; 2003. p. 650. doi:10.1201/9780203911983.ch7.
31. Lapprand A, Boisson F, Delolme F, Méchin F, Pascault JP.
Reactivity of isocyanates with urethanes: conditions for allo-
phanate formation. Polym Degrad Stab. 2005;90(2):363–73.
doi:10.1016/j.polymdegradstab.2005.01.045.
32. Delebecq E, Pascault J-P, Boutevin B, Ganachaud F. On the
versatility of urethane/urea bonds: reversibility, blocked iso-
cyanate, and non-isocyanate polyurethane. Chem Rev.
2013;113(1):80–118. doi:10.1021/cr300195n.
33. Wirpsza Z. Polyurethanes, chemistry, technology and applica-
tions. Polymer science and technology series. New York: Ellis
Horwood; 1993.
34. Sonnenschein MF. Introduction to polyurethane chemistry.
Polyurethanes. Wiley; 2014. p. 105–26.
35. Dušek K, Špı́rková M, Ilavský M. Network formation in poly-
urethanes due to allophanate and biuret formation: gel fraction
and equilibrium modulus. Makromol Chem Macromol Symp.
1991;45(1):87–95. doi:10.1002/masy.19910450112.
36. Semsarzadeh MA, Navarchian AH. Effects of NCO/OH ratio and
catalyst concentration on structure, thermal stability, and cross-
link density of poly(urethane-isocyanurate). J Appl Polym Sci.
2003;90(4):963–72.
37. Redakcji O, Ryszkowska J. Supermolecular structure, morphol-
ogy and physical properties of urea-urethane elastomers. Polim-
ery. 2012;57(11–12):777–85.
38. Dey J, Xu H, Shen J, Thevenot P, Gondi SR, Nguyen KT, et al.
Development of biodegradable crosslinked urethane-doped
polyester elastomers. Biomaterials. 2008;29(35):4637–49. doi:10.
1016/j.biomaterials.2008.08.020.
39. Thomson T, inventor Timothy Thomson, assignee. Hydrophilic
polyurethane as a matrix for the delivery of skin care ingredients.
US patent 20020182245. 2002.
40. Macosko CW, Miller DR. A new derivation of average molecular
weights of nonlinear polymers. Macromolecules.
1976;9(2):199–206. doi:10.1021/ma60050a003.
41. Petrović ZS, Zhang W, Javni I. Structure and properties
of polyurethanes prepared from triglyceride polyols by
ozonolysis. Biomacromol. 2005;6(2):713–9. doi:10.1021/bm049
451s.
42. Fox TG, Loshaek S. Influence of molecular weight and degree of
crosslinking on the specific volume and glass temperature of
polymers. J Polym Sci. 1955;15(80):371–90.
933
43. Clark AJ, Hoong SS. Copolymers of tetrahydrofuran and
epoxidized vegetable oils: application to elastomeric poly-
urethanes. Polym Chem. 2014;5(9):3238–44. doi:10.1039/c3py
01527k.
44. Menard KP. Dynamic mechanical analysis: a practical introduc-
tion. Boca Raton: CRC Press; 2008.
45. Kevin M. Thermomechanical and dynamic mechanical analysis.
In: Lobo H, Bonilla JV, editors. Handbook of plastics analysis.
Boca Raton: CRC Press, 2003. p. 102–90.
46. Zieleniewska M, Auguścik M, Prociak A, Rojek P, Ryszkowska
J. Polyurethane-urea substrates from rapeseed oil-based polyol
for bone tissue cultures intended for application in tissue engi-
neering. Polym Degrad Stab. 2014;. doi:10.1016/j.poly
mdegradstab.2014.03.010.
47. Puszka A, Kultys A. New thermoplastic polyurethane elastomers
based on aliphatic diisocyanate. J Therm Anal Calorim.
2017;128(1):407–16. doi:10.1007/s10973-016-5923-7.
48. Guan J, Sacks MS, Beckman EJ, Wagner WR. Biodegradable
poly(ether ester urethane)urea elastomers based on poly(ether
ester) triblock copolymers and putrescine: synthesis, characteri-
zation and cytocompatibility. Biomaterials. 2004;25(1):85–96.
doi:10.1016/S0142-9612(03)00476-9.
49. Hong Y, Guan J, Fujimoto KL, Hashizume R, Pelinescu AL,
Wagner WR. Tailoring the degradation kinetics of poly(ester
carbonate urethane)urea thermoplastic elastomers for tissue
engineering scaffolds. Biomaterials. 2010;31(15):4249–58.
doi:10.1016/j.biomaterials.2010.02.005.
50. Aung MM, Yaakob Z, Kamarudin S, Abdullah LC. Synthesis and
characterization of Jatropha (Jatropha curcas L.) oil-based
polyurethane wood adhesive. Ind Crops Prod. 2014;60:177–85.
doi:10.1016/j.indcrop.2014.05.038.
51. Javni I, Petrović ZS, Guo A, Fuller R. Thermal stability of
polyurethanes based on vegetable oils. J Appl Polym Sci.
2000;77(8):1723–34. doi:10.1002/1097-4628(20000822)77:
8\1723:aid-app9[3.0.co;2-k.
52. Beyler CL, Hirschler MM. Chapter 7, Thermal decomposition of
polymers. In: SFPE handbook of fire protection engineering, vol.
2. 2002. p. 111–31.
53. Hablot E, Zheng D, Bouquey M, Avérous L. Polyurethanes based
on castor oil: kinetics, chemical, mechanical and thermal prop-
erties. Macromol Mater Eng. 2008;293(11):922–9. doi:10.1002/
mame.200800185.
54. Prime RB, Bair HE, Vyazovkin S, Gallagher PK, Riga A. In:
Menczel JD, Prime RB, editors. Thermogravimetric analysis
(TGA). Thermal analysis of polymers. Wiley; 2009. p. 241–317.
doi:10.1002/9780470423837.
55. Kissinger HE. Reaction kinetics in differential thermal analysis.
Anal Chem. 1957;29(11):1702–6. doi:10.1021/ac60131a045.
56. Bogdanov B, Toncheva V, Schacht E, Finelli L, Sarti B, Scandola
M. Physical properties of poly(ester-urethanes) prepared from
different molar mass polycaprolactone-diols. Polymer.
1999;40(11):3171–82. doi:10.1016/S0032-3861(98)00552-7.
123