STRAIN-INDUCED CRYSTALLIZATION OF NATURAL RUBBER:

EFFECT OF PROTEINS AND PHOSPHOLIPIDS

SUREERUT AMNUAYPORNSRI, JITLADDA SAKDAPIPANICH*

DEPARTMENT OF CHEMISTRY AND CENTER OF EXCELLENCE FOR INNOVATION IN CHEMISTRY
FACULTY OF SCIENCE, MAHIDOL UNIVERSITY, BANGKOK 10400, THAILAND

SHIGEYUKI TOKI, BENJAMIN S. HSIAO

DEPARTMENT OF CHEMISTRY, STONY BROOK UNIVERSITY, NY 11794, USA

NAOYA ICHIKAWA
SUMITOMO RUBBER INDUSTRY R&D LTD. KOBE 651-0071, JAPAN

YASUYUKI TANAKA
INSTITUTE OF SCIENCE AND TECHNOLOGY FOR RESEARCH AND DEVELOPMENT, MAHIDOL
UNIVERSITY, SALAYA CAMPUS, NAKHONPATHOM 73170, THAILAND

ABSTRACT
The effects of proteins and phospholipids in natural rubber (NR) on the strain-induced crystallization behavior dur-
ing uniaxial deformation were studied by in-situ synchrotron wide-angle X-ray diffraction (WAXD) technique and simul-
taneous measurements of stress-strain relation. The influences of proteins and phospholipids in NR were evaluated sep-
arately by decomposition methods using deproteinization and lipase treatment, respectively. It was found that both com-
ponents form a naturally occurring network, which is responsible for the strain-induced crystallizability of unvulcanized
NR and the corresponding high mechanical property. This network also plays a significant role in strain-induced crys-
tallization of vulcanized natural rubber.

INTRODUCTION

Natural rubber (NR) from Hevea brasiliensis accounts for more than 99% of the world’s nat-
ural source of rubber, in which the major component is cis-1,4-polyisoprene. The plant has very
high productivity and the final product usually exhibits excellent properties, such as high tensile
and tear strength, good crack growth resistance and minimal heat buildup, often superior to syn-
thetic polyisoprene (IR). It is well known that even when IR contains isoprene units with cis-1,4
configuration higher than 98% (i.e., polymerized by Ti/Al catalyst) its mechanical properties are
still poorer than NR. Although the difference in mechanical property between NR and IR has
been attributed to the discrepancy in chemical composition, NR contains about 6% w/w of non-
rubber components, essentially proteins and phospholipids, whereas IR does not have any. The
exact role of proteins and phospholipids in NR in terms of the property enhancement is not clear,
which is the purpose of this study.
As reported previously,1 a linear rubber chain in NR consists of ω-terminal, two trans-1,4
isoprene units, 1,000-3,000 cis-1,4 isoprene units, and α-terminal. The α-terminal is composed
of mono- or di-phosphate group linked with phospholipids,2,3 which plays a role in branching
formation of rubber molecules.4 On the other hand, the ω-terminal of rubber molecule in NR was
postulated to be a modified dimethylallyl unit linked with a functional group, which can be asso-
ciated with proteins to form crosslinks through intermolecular hydrogen bonding.5 The proposed
structure of a linear rubber chain and naturally occurring network in NR are shown in Figure 1.

* Corresponding author. Ph: 662-889-331; Fax: 662-889-3116; email: scjtp@mahidol.ac.th.

753
754 RUBBER CHEMISTRY AND TECHNOLOGY VOL. 81

In this figure, the non-rubber components, composed of proteins and phospholipids, are pre-
sumed to be major constituents to form network points by interactions with rubber chains at both
chain-ends. These components with polar functional groups are immiscible with rubber chains.

FIG. 1. — Proposed structure of linear rubber chain and naturally occurring network in NR.

The effect of proteins and phospholipids on the tensile strength of unvulcanized NR was
investigated in a previous work.6-8 The removal of proteins in NR was found to show no change
in the tensile strength, but the procedure increased the strain at break.6 The decomposition of net-
work points through transesterification decreased the tensile strength of NR to a very low level,
comparable to that of typical IR.7,8 It is thought that the transesterification process decomposes
the network points containing ester linkages of phospholipids and phosphate group, thus render-
ing NR into linear chains of isoprene. Although the effect of proteins and phospholipids on ten-
sile strength of NR has been studied,6-8 little information on the relation with strain-induced crys-
tallization is available.
The mechanism of strain-induced crystallization is thought to be primarily responsible for
the outstanding tensile strength and good crack growth resistance in NR, where a great deal of
structure and property studies have been carried out in both vulcanized9-25 and unvulcanized
states.26-36 In these studies, the behavior of molecular orientation and strain-induced crystalliza-
tion of vulcanized NR have been well characterized by using 1H and 2H NMR (nuclear magnet-
ic resonance),9,10 infrared (IR) absorption,11 birefringence12-14 and wide-angle X-ray diffraction
(WAXD)15-25 techniques. In contrast, the changes in molecular orientation and strain-induced
crystallization of unvulcanized NR have been reported limitedly.26,33-36 Therefore, the purpose of
this work is to demonstrate changes of molecular orientation and strain-induced crystallization
during deformation of unvulcanized NR in relation with the effect of proteins and phospholipids
in natural rubber. Furthermore, the effect of these components on the molecular orientation and
strain-induced crystallization under the same deformation conditions in vulcanized NR was also
compared.
 STRAIN-INDUCED CRYSTALLIZATION OF NATURAL RUBBER 755

EXPERIMENTAL
SAMPLE PREPARATION

Natural rubber (NR) latex used in this study was from regularly tapped Hevea tree of RRIM
600 clone, provided by the Thai Rubber Latex Co., Thailand. The NR latex sample was cast on
a plate and dried in an oven at 50 oC for 24 h. Deproteinized NR (DPNR) was prepared by incu-
bation of NR latex (30% w/v dry rubber content) with 0.04 % w/v proteolytic enzyme (KAO KP-
3939) and 1% w/v Triton® X-100 for 12 h at 37 oC followed by centrifugation at 13,000 rpm for
30 min. The cream fraction was re-dispersed with 0.5% w/v Triton® X-100 to make 30% w/v dry
rubber content and re-centrifuged at 13,000 rpm for 30 min.
Lipase-treated DPNR (L-DPNR) was prepared using the following procedures to remove
neutral lipids and polar lipids, including phospholipids and glycolipids in NR2. The DPNR latex
(10% w/v dry rubber content) was mixed with lipase from microorganism (Candida rugosa,
Sigma) at 37 °C with pH 7.2 and incubated at 37 °C for 48 hr at the lipase: DPNR concentration
ratio of 5:1 by weight. The rubber fraction was recovered by centrifugation at 13,000 rpm for 30
min. The cream fraction was subsequently redispersed in 0.5% w/v Triton® X-100 solution and
treated with 0.1% w/v proteolytic enzyme for 12 h at 37 oC in order to eliminate the residual
lipase. The cream fraction was recovered again by centrifugation at 13,000 rpm for 30 min and
then it was subsequently redispersed in distilled water.
L-DPNR latex was used as a starting material for re-addition of proteins and phospholipids.
Proteins were extracted from fresh NR latex by the following procedures (i.e., washing with sur-
factant). Fresh NR latex was dispersed with 1% w/v Triton® X-100, followed by centrifugation
at 19,000 rpm for 30 min to separate serum from rubber fraction. The serum fraction was added
with 3 times volume of cold acetone and then the mixture was kept at 4 oC for 12 h in order to
precipitate the proteins in serum. The precipitated proteins were collected and re-dispersed in
0.5% w/v Triton® X-100 aqueous solution. Then, they were added back to L-DPNR latex in the
same ratio as the original latex and mixed by mechanical stirrer. Phosphatidyl choline (1 phr) was
used in this study as a representative of phospholipids in NR. It was also mixed with L-DPNR
latex by mechanical stirrer. These mixtures were preserved with 0.6% w/v NH3 solution for a
month before cast into a glass plate and dry in an oven at 50 oC for 24 h.
Transesterified NR (TENR) was prepared by the reaction of NR with freshly prepared sodi-
um methoxide in toluene solution at room temperature for 3 h, followed by precipitation using
an excess amount of methanol.7
NR, DPNR and L-DPNR were mixed with 2 phr dicumyl peroxide (DCP) in 4% w/v chlo-
roform solution. Then, the rubber mixture was cast and dried at room temperature to form 1 mm
thickness film. The received film was vulcanized at 150 oC for 30 min. The crosslink density of
vulcanized rubbers is shown in Table II.

CHARACTERIZATION

The nitrogen content of the sample was analyzed by a Leco Nitrogen Analyzer (model FP
528) with the sensitivity of 0.001%. The quantity of long-chain fatty acid ester group was deter-
mined by FTIR measurement based upon calibration by a mixture of methyl stearate and syn-
thetic cis-1,4-polyisoprene, Kuraprene IR10. The content of fatty acid ester group per weight of
rubber was determined by the intensity ratio of peaks at 1739 cm-1 (C=O) to 1664 cm-1 (C=C).37
The gel content was determined by toluene solubility measurement. The crosslink density
(υ) of vulcanized samples was estimated according to the equation from the theory of rubber
elasticity.38
756 RUBBER CHEMISTRY AND TECHNOLOGY VOL. 81

σ = υkT(α − α −2 ) (1)

where σ is the force per unit area, υ is the number of network chain in unit volume, k is the
Boltzmann constant, T is the absolute temperature, and α is the elongation ratio. A plot of σ ver-
sus (α – α-2) can yield a straight line and the value of υ can be calculated from the initial slope.

IN-SITU WAXD MEASUREMENT DURING TENSILE DEFORMATION

The behavior of strain-induced crystallization in the chosen NR samples was characterized

by using in-situ synchrotron WAXD technique at the X27C beamline in the National Synchrotron
Light Source (NSLS), Brookhaven National Laboratory (BNL). The wavelength used at X27C
was 0.1371 nm. The WAXD patterns were recorded by a MAR-CCD X-ray detector. The expo-
sure time for each image was 30 sec. The tensile machine for X-ray measurement was a tabletop
stretching machine from the Instron Company, the machine has been modified specifically for X-
ray study in a symmetric deformation mode, allowing the X-ray beam to illuminate the same
sample position during straining. The sample was stretched at a speed of 10 mm/min. The orig-
inal sample length was 30 mm. The strain was calculated by (l-lo)/lo where l is the sample length
during stretching and lo is the original sample length. The stress was calculated by F/Ao, where
F is a measured force and Ao is the original cross section of the sample. The experiment was car-
ried out at 25 oC.
The collected WAXD pattern consists of contributions mainly from isotropic, oriented amor-
phous and oriented crystalline components. The following procedures were used to evaluate each
contribution. As describe in our previous paper,19,20 after subtraction of background scattering
and correction of beam fluctuation, the isotropic amorphous component was subtracted referring
to the WAXD pattern of un-stretched sample. The rest of the pattern contains both oriented amor-
phous and oriented crystalline fractions. An integrated intensity profile was calculated.
Subsequently, a peak fitting program was used to decompose the integrated intensity profile into
orientated amorphous and oriented crystalline components. The calculation of % anisotropic, %
oriented amorphous and % crystalline fraction can be described as follows:

% Crystalline fraction = Integrated scattering intensity of crystalline

(2)
components x 100 / Integrated scattering intensity of all components

% Oriented amorphous fraction = Integrated scattering intensity of orientation

(3)
components x 100 / Integrated scattering intensity of all components

% Anisotropic fraction = % Crystalline fraction + % Oriented

(4)
amorphous fraction

RESULTS AND DISCUSSION

The roles of proteins and phospholipids were investigated by composition alternations
through deproteinization and lipase treatments, respectively. As shown in Table I, the nitrogen
content of NR, which represents the amount of proteins, decreased to almost 0% after depro-
teinization. The effectiveness of lipase treatment for DPNR was also confirmed by the analysis
of the ester content in long-chain fatty acid. The ester content decreased from 25.6 to 16.2
mmol/kg rubber after lipase treatment, indicating that some neutral lipids and polar lipids,
including phospholipids, were removed. It should be noted that the efficiency of lipase may not
be enough to decompose all lipids. In contrast, the transesterification process can decompose and
 STRAIN-INDUCED CRYSTALLIZATION OF NATURAL RUBBER 757

remove almost all proteins and lipids. Since proteins and phospholipids are associated with the
formation of naturally occurring network in raw NR, the decomposition of proteins and/or phos-
pholipids is expected to affect the integrity of the network structure, which was evidenced by the
decrease of gel content.

TABLE I
CHARACTERIZATION OF UNVULCANIZED RUBBERS
Sample Nitrogen content Ester content Gel content
(%w/w) (mmol/kg-rubber) (%w/w)
NR 0.75 25.6 23.9
DPNR 0.01 25.8 5.15
L-DPNR 0.01 16.2 0
TENR 0.02 0 0

Stress-strain curves of unvulcanized NR, DPNR, L-DPNR and TENR are shown in Figure
2. It is seen that the stress of NR increased significantly at the onset strain of 3.0, while the onset
strain of DPNR appears at strains of 4.0. L-DPNR does not show the onset strain clearly even it
was stretched until the strain of 6.0. TENR could not be extended more than strain of 4.0 and did
not show any increase in stress beyond the strain of 1.0 during stretching, which can be explained
by the absence of network structure in TENR. In general, NR showed higher stress than DPNR
and L-DPNR in all range of deformation. The removal of proteins and phospholipids clearly
destroys the naturally occurring network, resulting in significant decreases in tensile properties.
During retraction, NR, DPNR and L-DPNR samples recovered and showed clear hysteresis with
permanent sets of 130%, 150% and 170%, respectively. It appears that the decomposition of pro-
teins and phospholipids not only decreases the modulus but also increases the permanent set
because of the destruction of network points.

FIG. 2. — Stress-strain curves of raw NR, DPNR, L-DPNR and TENR during stretching and retraction.
758 RUBBER CHEMISTRY AND TECHNOLOGY VOL. 81

The WAXD patterns collected at strain of 6.0 for NR, DPNR, L-DPNR samples and at strain
of 4.0 for TENR sample are shown in Figure 3. The stretched NR, DPNR and L-DPNR samples
all exhibited the oriented 120, 200 and 201 crystalline reflections superimposed with an isotrop-
ic scattering halo. However, the stretched TENR sample only showed an isotropic scattering halo
from random amorphous chains without any indication of fiber diffraction pattern. The peculiar
characteristic of TENR is ascribed to the complete decomposition of naturally occurring network
through transesterification.

FIG. 3. — WAXD patterns of (a) NR; (b) DPNR; (c) L-DPNR at strain of 6.0; and (d) TENR at strain of 3.8.

Changes in the anisotropic, oriented amorphous and crystalline fraction during extension
and retraction of NR, DPNR and L-DPNR samples are shown in Figure 4. The anisotropic frac-
tion consists of orientated amorphous and oriented crystalline phases. The anisotropic fraction
was detected during stretching at the strain of about 2.0 for NR and DPNR, while at about 3.0
for L-DPNR. All samples showed increases of the anisotropic fraction during stretching. It is
interesting to note that the maximum percentage of the anisotropic fraction at high strains (e.g.
strain of 6.0) was less than 10%, indicating that the majority of chains (>90%) remain in the un-
orientated amorphous state even under substantial deformation. The increase in the anisotropic
fraction of NR was faster than those of DPNR and L-DPNR; at a given strain, NR also showed
a higher value of anisotropic fraction than DPNR and L-DPNR. These findings indicate that the
naturally occurring network, originated by the interaction of proteins and phospholipids with
rubber chain ends, play a significant role in the orientation of isoprene molecules and the conse-
quential strain-induced crystallization during deformation.
 STRAIN-INDUCED CRYSTALLIZATION OF NATURAL RUBBER 759

The oriented amorphous fraction during stretching was first observed at strain of about 2.0
for NR and DPNR, while it was of 3.0 for L-DPNR. It was found that the fraction of oriented
amorphous chains in unvulcanized rubber was lower than 2%. The crystalline fraction was first
detected at strain of 3.0 for NR and DPNR and at strain of 4.0 for L-DPNR. It was seen that at
a given strain, the fraction of strain-induced crystal was suppressed after the removal of proteins
(through deproteinization). Furthermore, the removal of proteins and phospholipids by the com-
bined deproteinization and lipase treatments resulted in very low fraction of strain-induced crys-
tal. It is note that the removal of proteins and phospholipids resulted in the decomposition of the
naturally occurring network in NR. Thus, the decrement of strain-induced crystalline fraction
after the removal of proteins and phospholipids implies that the naturally occurring network is
responsible for the strain-induced crystallizability of unvulcanized NR. It could be confirmed by
no crystalline reflection pattern was observed in TENR during stretching.

FIG. 4. — Change in the anisotropic fraction, the oriented amorphous fraction and crystalline fraction
of NR, DPNR and L-DPNR during extension and retraction.

Effects of proteins and phospholipids on the behavior of strain-induced crystallization were

further explored with the following experiments. The components of proteins and phospholipids
were added back to the L-DPNR latex. The first sample preparation involved the addition of pro-
teins (which were extracted from fresh NR latex by washing with surfactant) in L-DPNR to the
same protein concentration as in the original latex. The second sample preparation involved the
addition of phosphatidyl choline (1 phr), which is representative of phospholipids in NR, in L-
DPNR to the same phospholipid concentration as in the original latex. The third sample prepa-
ration involved the addition of both proteins and phosphatidyl choline to the appropriate level of
original concentrations. In this experiment, the rubber samples were stretched until break in
order to gain more the crystalline fraction of L-DPNR, which showed very low value even at
strain of 6.0 as shown in Figure 4.
760 RUBBER CHEMISTRY AND TECHNOLOGY VOL. 81

The stress-strain curves of pure L-DPNR and samples after additions of extracted proteins
and/or phosphatidyl choline are shown in Figure 5. It is interesting to see that additions of
extracted proteins, phosphatidyl choline and both of proteins and phosphatidyl choline to L-
DPNR did not improve the tensile properties. This suggests that high tensile strength of NR is
not directly caused by the compositions of proteins and/or phospholipids alone, but it is related
to the characteristic network structure in NR, which originates from the bonding of proteins and
phospholipids with the terminal units of isoprene chains. It is clear that the simple addition of
extracted proteins and/or phosphatidyl choline is not sufficient to re-construct the naturally
occurring network in NR.

FIG. 5 . — Stress-strain curves of L-DPNR added with extracted proteins and/or phosphatidyl choline.

WAXD patterns of pure L-DPNR and L-DPNR with additions of extracted proteins and/or
phosphatidyl choline at strain of 7.5 are shown in Figure 6. Although all samples exhibited dis-
tinct oriented fiber diffraction patterns, the diffraction pattern in L-DPNR with proteins and/or
phosphatidyl choline showed a lower level of intensity than that in original L-DPNR. This indi-
cates that the simple addition of extracted proteins and/or phosphatidyl choline decreases the
crystallizability of rubber chains.
 STRAIN-INDUCED CRYSTALLIZATION OF NATURAL RUBBER 761

FIG. 6. — WAXD patterns of (a) L-DPNR; (b) L-DPNR added with extracted proteins; (c) L-DPNR added with
phosphatidyl choline; and (d) L-DPNR added with extracted proteins and phosphatidyl choline at strain of 7.5.

Changes in the crystalline fraction of L-DPNR after the addition of proteins and/or phos-
phatidyl choline are shown in Figure 7. The strain-induced crystallization in L-DPNR appeared
at strain of about 4.0, whereas that in L-DPNR added with proteins and/or phosphatidyl choline
appeared at strain of about 5.0. The strain-induced crystalline fraction in L-DPNR was clearly
suppressed by the addition of extracted proteins and/or phosphatidyl choline. Although the crys-
talline fraction of L-DPNR added with phosphatidyl choline increases with strain close to L-
DPNR, the ones of L-DPNR added with protein and L-DPNR added with both of protein and
phosphatidyl choline does not increase significantly. The behavior of crystallization in four sam-
ples is similar to the one of the stress-strain relations in them as shown in Figure 5. The un-
improvement of stress-strain behavior and strain-induced crystallization after the re-addition of
extracted proteins and phosphatidyl choline suggests that the superior properties are not directly
caused by the chemical presence of proteins and phospholipids but their interactions with rubber
chain ends to form the network structure.
762 RUBBER CHEMISTRY AND TECHNOLOGY VOL. 81

FIG. 7. — Change in the crystalline fraction of L-DPNR after the addition

of extracted proteins and/or phosphatidyl choline.

We also made an attempt to investigate the effects of proteins and phospholipids on strain-
induced crystallization of different rubbers (NR, DPNR and L-DPNR) in vulcanized state. To
maintain the naturally occurring network, formed by proteins and phospholipids in raw rubber,
we avoided the usage of internal mixer and two-roll mill. Instead, the process of peroxide vul-
canization was carried out in chloroform. With this process, the crosslink density of resulting
vulcanized rubber is very low as shown in Table II.

TABLE II
CROSSLINK DENSITY OF PEROXIDE VULCANIZED RUBBER
Vulcanized rubber Crosslink density x 104 (mol/cm3)
V-NR 0.97
V-DPNR 0.80
V-LDPNR 0.66

The stress-strain curves of vulcanized NR, DPNR and L-DPNR are shown in Figure 8. It is
seen that vulcanized NR exhibited the highest level of stress compared with those of vulcanized
DPNR and vulcanized L-DPNR at all strains. This indicates that the naturally occurring network,
formed by proteins and phospholipids, also affect the tensile properties in the vulcanized state.
During retraction, all vulcanized rubbers recovered and showed hysteresis with permanent sets
of 100%.
 STRAIN-INDUCED CRYSTALLIZATION OF NATURAL RUBBER 763

FIG. 8. — Stress-strain curves of vulcanized NR, DPNR and L-DPNR during extension and retraction.

FIG. 9. — WAXD patterns of (a) vulcanized NR; (b) vulcanized DPNR; and (c) vulcanized L-DPNR at strain of 6.0.

WAXD patterns of vulcanized NR, vulcanized DPNR and vulcanized L-DPNR samples,
collected at strain of 6.0 are shown in Figure 9. All stretched vulcanized rubbers showed distinct
120, 200 and 201 crystalline reflections superimposed with an isotropic scattering halo, the same
764 RUBBER CHEMISTRY AND TECHNOLOGY VOL. 81

feature as unvulcanized rubbers. Since the intensities of crystalline reflections in all vulcanized
samples were almost the same, the peroxide crosslink network predominantly governed the
strain-induced crystallization of vulcanized NR, DPNR and L-DPNR.
Changes in the anisotropic, oriented amorphous and crystalline fraction of vulcanized rub-
bers (NR, DPNR and L-DPNR) during extension and retraction are shown in Figure 10. The
anisotropic fraction was detected at strain of about 2.0 for all samples and it increased during
stretching. The anisotropic fraction of vulcanized NR was higher than those of vulcanized DPNR
and vulcanized L-DPNR at lower strain than 6.0. However, the anisotropic fractions of all vul-
canized rubbers at strain of 6.0 are about the same. The oriented amorphous fractions of all vul-
canized rubbers appear at strain of about 2.0 and the maximum attainable values at high strains
are always lower than 3%.

FIG. 10. — Change in the anisotropic, oriented amorphous and crystalline

fraction of vulcanized NR; DPNR and L-DPNR during extension and retraction.

Corresponding changes in the crystalline fraction of vulcanized rubbers (NR, DPNR and L-
DPNR) during extension and retraction are also determined as shown in Figure 10. The strain-
induced crystallization appeared at strain of about 3.0 for vulcanized NR and vulcanized DPNR,
while at strain of about 4.0 for vulcanized L-DPNR. It is clear that the onset strain for crystal-
lization in NR, DPNR and L-DPNR is unaltered after vulcanization. In other words, L-DPNR
exhibit the retarded behavior of strain-induced crystallization, compared to NR and DPNR in
both unvulcanized and vulcanized states. At a given strain, the crystalline fraction of vulcanized
NR is slightly higher than vulcanized DPNR and vulcanized L-DPNR. This implies that the nat-
urally occurring network, formed by proteins and phospholipids, contributes to strain-induced
crystallization even in the vulcanized state. The crystalline fraction at strain of 6.0 was almost
the same in all samples. We note that the strain-induced crystallization of vulcanized rubber is
 STRAIN-INDUCED CRYSTALLIZATION OF NATURAL RUBBER 765

not simply described by not only the naturally occurring network but also the chemical network.

CONCLUSIONS
The naturally occurring network, formed by proteins and phospholipids, is responsible for
the high tensile strength and crystallizability under deformation in unvulcanized NR. The
removal of proteins and phospholipids by deproteinization and lipase treatments, respectively,
resulted in a significantly decrease in stress-strain behavior and the crystallizability under defor-
mation. The addition of extracted proteins and/or phosphatidyl choline to L-DPNR (depro-
teinized and lipase-treated NR) was not effective to improve the stress-strain behavior and the
crystallizability upon stretching, since the superior properties are not directly caused by the
chemical presence of proteins and phospholipids but their interactions with rubber chain ends to
form the network structure. The naturally occurring network, formed by interactions of proteins
and phospholipids with the terminal units of isoprene chains, is mainly responsible for the supe-
rior stress-strain behavior and the strain-induced crystallization of NR in both unvulcanized and
vulcanized states.

ACKNOWLEDGEMENT
The financial support of this study was provided by a grant from the Thailand Research Fund
PHD/0142/2546, the Center of Excellence for Innovation in Chemistry (PERCH-CIC), the
Commission on Higher Education RMU4980046 and the National Science Foundation (DMR-
0405432) with a special creativity extension award, Yokohama Rubber, SRI R&D and
Bridgestone.

REFERENCES
1Y. Tanaka, S. Kawahara, J. Tangpakdee, Kautsch. Gummi Kunstst. 50, 6 (1997).
2L. Tarachiwin, J. Sakdapipanich, K. Ute, T. Kitayama, T. Bamba, E. Fukusaki, A. Kobayashi, Y. Tanaka,
Biomacromolecules 6, 1851 (2005).
3L. Tarachiwin, J. Sakdapipanich, K. Ute, T. Kitayama, Y. Tanaka, Biomacromolecules 6, 1858 (2005).
4L. Tarachiwin, J. Sakdapipanich, Y. Tanaka, Kautsch. Gummi Kunstst. 3, 115 (2005).
5D. Mekkriengkrai, J.T. Sakdapipanich, T. Yasuyuki, RUBBER CHEM. TECHNOL. 79, 366 (2006).
6N. Ichikawa, A.H. Eng, Y. Tanaka, Proc. Int. Rubber Conf. Kuala Lumpur, 1993.
7A. H. Eng, S. Ejiri, S. Kawahara, Y. Tanaka, J. Appl. Polym Sci. Appl Polym. Symp. 53, 5 (1994).
8S. Kawahara, Y. Isono, T. Kakubo, Y. Tanaka, A. H. Eng, RUBBER CHEM. TECHNOL. 73, 39 (2000).
9J. Rault, J. Marchal, P. Judeinstein, P. A. Albouy, Eur. Phys. J. E. 21(3), 243 (2006).
10J. Rault, J. Marchal, P. Judeinstein, P. A. Albouy, Macromolecule 39, 8356 (2006).
11H.W. Siesler, Applied Spectroscopy 39, 761 (1985).
12S. Toki, T.Z. Sen, D. Valladares, M. Cakmak, “True Stress, True Strain And Real Time Birefringence Development Of
Orientation In Natural Rubber During Uniaxial Stretching As Detected By Spectral Birefringence Technique,” Paper
#12, presented at the meeting of the Rubber Division, ACS, Providence, Rhode Island, April 24-27, 2001.
13D. Valladares, S. Toki, T. Z. Sen, B. Yalcin, M. Cakmak, Macromol. Symp. 185, 149 (2002).
14I. S. Choi, C. M. Roland, RUBBER CHEM. TECHNOL. 70, 202 (1997).
15Y. Shimomura, J. L. White, J. Appl. Polym. Sci. 27, 3553 (1982).
16G. R. Mitchell, Polymer 25, 1562 (1984).
17S. Toki, T. Fujimaki, M. Okuyama, Polymer 41, 5423 (2000).
18S. Murakami, K. Senoo, S. Toki, S. Kohjiya, Polymer 43, 2117 (2002).
19S. Toki, I. Sics, S. Ran, L. Liu, B. S. Hsiao, S. Murakami, K. Senoo, S. Kohjiya, Macromolecules 35, 6578 (2002).
766 RUBBER CHEMISTRY AND TECHNOLOGY VOL. 81

20S. Toki, I. Sics, S. Ran, L. Liu, B. S. Hsiao, S. Murakami, K. Senoo, S. Kohjiya, Polymer 44, 6003 (2003).
21S. Toki, B. S. Hsiao, Macromolecules 36, 5915 (2003).
22M. Tosaka, S. Murakami, S. Poompradub, S. Kohjiya, Y. Ikeda, S. Toki, I. Sics, B. S. Hsiao, Macromolecules 37, 3299
(2004).
23S. Poompradub, M. Tosaka, S. Kohjiya, Y. Ikeda, S. Toki, I. Sics, B. S. Hsiao, J. Appl. Phys. 97, 103529 (2005).
24M. Tosaka, K. Senoo, S. Kohjiya, Y. Ikeda, J. Appl. Phys. 101, 084909 (2007).
25S. Kohjiya, M. Tosaka, M. Furutani, Y. Ikeda, S. Toki, B. S. Hsiao, Polymer 48, 3801 (2007).
26G. L. Clark, E. Wolthuis, W. H. Smith, J. Research Natl. Bur. Standards 19, 479 (1937).
27D. Luch, G. S. Y. Yeh, J. Macromol. Sci. Phys. B7, 121 (1973).
28B. C. Edwards, J. Polym. Sci. Polym. Phys. Ed. 13, 1387 (1975).
29E. Von Meerwal, R. B. Creel, D. F. Griffin, E. DiCato, F. T. Lin, L. M. Lin, J. Appl. Polym. Sci. 21, 1489 (1977).
30Y. Shimomura, J. L. White, J. E. Spruiell, J. Appl. Polym. Sci. 27, 3553 (1982).
31A. Cameron, W. J. Mcgill, J. Polym. Sci. Part A 27, 1071 (1989).
32G. D. Wathen, A. N. Gent, J. Nat. Rubb. Res. 5(3), 178 (1990).
33N. Rahman, A. Isanasari, R. Anggraeni, S. Honggokusumo, M. Iguchi, T. Masuko, K. Tashiro, Polymer 44, 283 (2003).
34M. M. Alam, T. Asano, J. Macromol. Sci. Part B Phys. 45, 753 (2006).
35S. Toki, I. Sics, B. S. Hsiao, S. Amnuaypornsri, S. Kawahara, “Strain-Induced Crystallization In Un-Vulcanized Natural
Rubbers By Synchrotron X-Ray Study,” Paper #75, presented at the meeting of the Rubber Division, ACS, Pittsburgh,
PA, November 1-3, 2005.
36S. Amnuaypornsri, S. Toki, B. S. Hsiao, J. Sakdapipanich, Y. Tanaka, “Strain-Induced Crystallization Of Natural
Rubber: Effect Of Proteins And Phospholipids,” Paper #16, presented at the meeting of the Rubber Division, ACS,
Cleveland, Ohio, October 16-18, 2007.
37A. H. Eng, J. Tangpakdee, S. Kawahara, Y. Tanaka, J. Nat. Rubb. Res. 12(1), 11 (1997).
38L. R. G. Treloar, “The Physics of Rubber Elasticity,” Oxford University Press, London (1975).

[ Paper 16, presented at the Fall Rubber Division, ACS, Meeting

(Cleveland, OH) October 16-18, 2007, revised May 2008 ]