J Polym Res (2012) 19:9864

DOI 10.1007/s10965-012-9864-y

ORIGINAL PAPER

Improvement of the properties of natural rubber/ground tire

rubber composites through biological desulfurization of GTR
Yuanhu Li & Suhe Zhao & Yaqin Wang

Received: 20 September 2011 / Accepted: 30 March 2012 / Published online: 17 April 2012
# Springer Science+Business Media B.V. 2012

Abstract The surface desulfurization of ground tire rubber return to the ecological environment through natural degrada-
(GTR) had been carried out via a biological treatment by tion because rubber vulcanizates have a stable cross-linked
Thiobacillus sp. with strong sulfur oxidizing capacity. The three-dimensional network structure [1, 2]. The reuse or recy-
bonding states and element content on the surface of GTR cling of waste rubber tires is a significant and word-wide
and desulfurated GTR (DGTR) were evaluated using an X-ray problem.
Photoelectron Spectroscopy (XPS). The contact angle of GTR A potentially attractive recycling or utilizing way is to
was 120.5° and decreased down to 93.5° after treatment. The grind waste tire rubber into ground tire rubber (GTR) and use
cure characteristics, swelling behavior, and crosslink density the GTR as a compounding ingredient or as a replacement of
of natural rubber (NR)/GTR and NR/DGTR were examined. raw rubber. The reuse of GTR not only achieves the goals of
The improvement in mechanical properties was observed for protecting the environment but also saves limited petroleum
NR/DGTR vulcanizates, which was attributed to the enhanced resources from which the raw material is originated. How-
interfacial interaction between DGTR and NR matrix. The ever, the bonds between GTR and the rubber matrix are very
dynamic mechanical analysis (DMA) results showed that weak because of the crosslink structure of GTR and lack of
NR/DGTR vulcanizates had a reduction of molecular chain reactive sites on the GTR surface, resulting in a decrease of
friction resistance during glass transition region and Scanning the mechanical properties of composites. One possible way
electron microscopy (SEM) studies further indicated a good to overcome this problem is to break the sulfur crosslink
coherency and homogeneity between DGTR and NR matrix. bonds on the surface of GTR, thereby make the hydrocarbon
chain regain flexible and increase the number of reactive
Keywords Ground tire rubber . Desulfurization . Natural sites. The crosslink bonds of ground rubber can be cleaved
rubber . Composites . Mechanical properties via numerous techniques, primarily physical and chemical
methods [3–6]. Both methods are efficient to enhance the
bonds between the ground rubber and matrix [7–10], but the
Introduction methods have high energy consumption or secondary pol-
lution involved.
The large quantities of worn out tires have posed challenging In recent years, several microorganisms that exhibit bio-
environmental and disposal problems. These materials cannot logical activity towards sulfur have been used for the desul-
furization of ground rubber [11–14]. These microorganisms
Y. Li : S. Zhao (*) can decrease the sulfur content or oxidize sulfur to sulfate,
Key Laboratory of Beijing City on Preparation and Processing increasing the reactive sites on the surface of ground rubber.
of Novel Polymer Materials, College of Materials Science
Microbial desulfurization processes have several advantages,
and Engineering, Beijing University of Chemical Technology,
Beijing 100029, China breaking only the surface sulfur cross-linked bond of ground
e-mail: [email protected] rubber, cost-effective, and environmentally friendly, com-
pared with physical and chemical processes which normally
Y. Wang
are energy demanding or require hazardous chemicals.
College of Life Science and Technology,
Beijing University of Chemical Technology, This article attempted to find a cost-effective and environ-
Beijing 100029, China mental friendly technique for recycling waste rubber by
Page 2 of 7 J Polym Res (2012) 19:9864

biological treatment. Thiobacillus sp. had been reported that it Table 1 Formulation of
NR composites Material Weight/phr
could decrease the sulfur content of GTR and oxidize sulfur to
sulfate [15, 16]. In this study, Thiobacillus sp. was used for NR 100
biological surface desulfurization of ground tire rubber. The Zinc oxide 4
bonding states, element content, and contact angle measure- Stearic acid 2
ment were characterized to evaluate the surface of GTR after Carbon Black N330 30
treatment. The cure characteristics, crosslink density, mechan- Accelerator DPG 0.6
ical properties, and dynamic mechanical analysis of NR/GTR Accelerator MBTS 1.2
and NR/DGTR were examined. SEM studies of tensile frac- Sulfur 2
ture surface were used to evaluate interface and coherency GTR or DGTR Variable (0,10,
between DGTR and NR matrix. 20, 30 or 40)

Experimental Characterization and measurements

Materials GTR size distribution studies

Natural rubber was supplied by Xishuang Banna Eastwind The size distribution of GTR was analyzed using a laser particle
Farmland, China. The GTR used in this study was supplied size distribution analyzer (LS-Pop3, OMEC Technology Co.,
by Puyang Rubber Factory, Henan Province, China. Carbon Ltd., Zhuhai, China).
Black N330 was obtained from Tianjin Dolphin Carbon
Black Development Co. Thiobacillus sp. strain was isolated Microorganism biomass
from the soil samples of sulfur-rich coal mine in Sichuan
province, China. The Growth of the microorganism in the medium was moni-
tored by optical density at 600 nm (XSZ-H3 binocular bio-
logical microscope, Chongqing Optical Instrument Factory,
Biological desulfurization process
China).
The compounds of the medium were (g/L): KH2PO4, 4.0;
XPS analysis
K2HPO4·3H2O, 4.0; MgSO4·7H2O, 0.8; NH4Cl, 0.4; CaCl2,
0.01, and Na2S2O3·5H2O, 10.0. The medium had been auto-
XPS measurement was carried out using a Thermo Electron
claved at 115 °C before the incubation and the cultivation
EscalabI250 instrument to evaluate the elements contents
condition was at 30 °C and pH at 6.5, with stirring at 175 rpm.
and bonding states on the surface of GTR before and after
The inoculum of Thiobacillus sp. strain was 10 % (v/v).
biological modification. Binding energies were corrected to
Before desulfurization, GTR was immerged in 75 % ethanol
the carbon 1 s peak located at 285.0 eV.
(v/v) for 24 h in order to kill the microorganisms attaching on
it and to remove harmful additives. After 3 days’ incubation,
GTR was added into culture medium. After 20 days’ desul- Water contact angle measurement
furization, DGTR was filtered out and washed by distilled
water for an hour. One method to appraise surface properties was the water
contact angle measurement. When water was applied to the
surface, the outmost surface layers interacted with the water.
Preparation of rubber compounds Hydrophobic surface with low free energy gave a high contact
angle with water and wet surface with high-energy allowed
Raw NR was masticated on the two-roll mill and then the drop to spread that gave a low contact angle. The mea-
blended with other additives and various contents of GTR surement of the contact angle of water was carried out at 25 °C
or DGTR for 10 min. The basic formulation was listed in (OCA15EC, dataphysics, Germany) and the values for the
Table 1. Rubber compounds were vulcanized on a platen contact angles were averaged over 5 different water drops.
press at 15 MPa and 150 °C according to ASTM D 2084–
07. Optimum cure time was determined from an Oscillating Sol fraction measurement
Disk Rheometer. Dumbbell-shaped tensile specimens were
punched out from the compression molded sheets along the The samples were trapped in a copper net and immersed in
mill grain direction. acetone for 24 h so as to extract polar substrates and then dried
J Polym Res (2012) 19:9864 Page 3 of 7

in a vacuum oven at 50 °C. The sol fractions before and after Result and discussion
desulfurization were obtained by 72 h Soxhlet extraction
method using toluene as a solvent. The weights of samples Morphology and particle size distribution of GTR
were recorded before and after the extraction.
Figure 1 showed the particle size distribution of the GTR
used in this work. Most of the GTR particles were smaller
Determination of crosslink density
than 100 μm, and 74 % of the particles were in the range
30 μm to 75 μm, mainly concentrated at 50 μm.
The determination of crosslink density of the vulcanized
rubber compounds were carried out using a swelling meth-
Growth of microorganisms
od. A test specimen, about 0.05 g, was immersed in 100 mL
of toluene at a controlled temperature of 30 °C for 72 h.
The growth curve of microorganisms (Fig. 2) showed that in
Excess solvent on the surface of the specimen was removed
the first 3 days of incubation, the microorganisms grew up
by blotting with filter paper. The weights of swollen speci-
naturally with low biomass (O.D. 600 nm). GTR and glucose
mens were measured. The crosslink densities of the rubber
were added into culture medium at the 3rd day and desulfur-
vulcanizates were determined by the Flory–Rehner equation
ization was allowed to proceed for the next 20 days. From 3rd
[17, 18]:
day to 5th day, the biomass almost remained unchanged,
because the residual chemicals in GTR prevented Thiobacillus
1nð1  v2 Þ þ v2 þ cv22 sp. growing. After 5th day, Thiobacillus sp. was adapted to the
ve ¼   1  ð1Þ environment, so biomass rapidly increased. The biomass
Vs v23  0:5v2 reached a maximum value at the 10th day, and then gradually
decreased because of consumption of nutrients.

Where ve was the crosslink density, Vs was the molar Element content and bonding states analysis
volume of the swelling solvent, v2 was the volume frac-
tion of toluene, and χ was the polymer solvent interaction XPS analysis was used to detect the variation of chemical
parameter. The interaction parameter for NR-toluene sys- composition of the GTR surface after treatment. Surface com-
tem was 0.39. positions determined from XPS analysis of GTR before and
after desulfurization were summarized in Table 2. The result
Mechanical properties studies showed that oxygen content on the surface of DGTR in-
creased by 30 %. The oxidization reaction might be catalyzed
Mechanical properties were measured at room temperature by enzymes in the microorganisms, improving the surface
according to ASTM D412 (tensile strength) and ASTM activity and reprocessing ability of desulfurated rubber.
D624 (tear strength). Mechanical data were averaged over Structural changes in sulfide bonds on the surface of
four specimens. GTR before and after treatment were shown in Fig. 3. Three
second splits of S 2p peaks at 162.3, 163.7, and 165.7 eV
were assigned as the bonding energy for S-C, S-S, and S-O
DMA analysis

The dynamic mechanical properties of NR composites were

measured by using a 01 dB-Metravib VA3000, France. A test
specimen for DMA was cut from molded sheets approximate-
ly 2.0 mm thick. The tests were run in tension mode at a fixed
frequency of 1 Hz and strain amplitude of 1 %. The temper-
ature range was from −100 °C to 100 °C, and the heating rate
was 3 °C/min.

Scanning electron microscopy

The morphology of tensile fracture surface of NR/DGTR

and NR/GTR were observed with a Hitachi S-4700 SEM,
Japan. The fracture surfaces of specimen were vacuum-
plated with gold for electrical conduction. Fig. 1 Particle size distribution of GTR
Page 4 of 7 J Polym Res (2012) 19:9864

Fig. 3 Structural changes in sulfide bonds on the surface of GTR

Fig. 2 Growth of microorganisms during cultivation (GTR was added before and after treatment
on the 3rd day)

crosslink scission occurred in GTR, the sol fraction was low

bonds, respectively [19]. The formation of S-O bonds was because the microbial desulfurization only happened on GTR
not observed on the surface of GTR. After treatment, it was surface down to a depth of several micrometers, and the inner
found that the frequency of S-S bonds and S-C bonds were parts of the GTR remained unchanged.
decreased by 18.3 % and 42.3 % respectively and the
formation of S-O bonds was observed on the surface of Cure characteristics
DGTR. The result showed that cross-linked sulfur bonds
were partly cleaved and converted to sulfate or oxygen- The effects of GTR and DGTR loading on the cure charac-
containing sulfur-based groups. So it resulted in an increase teristics were shown in Fig. 5. From Fig. 5, the minimum
in oxygen content and a decrease in sulfur content. The torque changed slightly with the increase of DGTR content,
introduction of oxygen containing functional groups on the but rose with increasing GTR content. This indicated that it
surface of GTR corresponded to the increase in surface was more difficult to process compounds containing GTR
energy. compared with compounds containing DGTR. The reason
was that cross-linked structure of GTR restricted the move-
Wettability of GTR and DGTR ment of molecular chains and adversely affected reprocessing

Water contact angle measurement was used to appraise the

wettability of GTR surface before and after treatment. From
Fig. 4, the contact angle of GTR was 120.5°, which revealed
the hydrophobic character of GTR. The water contact angle
of DGTR decreased down to 93.5°, implied that DGTR was
much more hydrophilic than GTR due to the oxidization
desulfurization reaction on the surface of DGTR.

Sol fraction

The desulfurization degree of GTR was evaluated by soluble

fraction. The sol fraction of GTR was 4.69 % and rose to
7.63 % after desulfurization, increased by 63 %. Even though

Table 2 Element con-

tent of GTR before and Element GTR DGTR
after treatment as
determined by XPS C (%) 96.41 95.79
O (%) 2.63 3.42
S (%) 0.96 0.79
Fig. 4 Water contact angle measurement of GTR and DGTR
J Polym Res (2012) 19:9864 Page 5 of 7

because sulfur migrated from matrix to particles in the

revulcanization process due to the higher sulfur concentra-
tion of gum NR, result in a decline in the crosslink density
of NR composites. Compared with NR/GTR vulcanizates,
crosslink density of NR/DGTR vulcanizates was lower,
because of the migration of sulfur being obstructed by
GTR with crosslink structure. After desulfurization, sulfur
content and cross-linked structure on DGTR surface was
reduced, so sulfur migrating from matrix to DGTR became
much easier.

Mechanical properties

The tensile strength, elongation at break, and tear strength

were used to evaluate the mechanical properties of the
Fig. 5 Minimum torque and maximum torque vs. waste rubber load-
ing in the NR matrix
systems. Generally, materials tended to become weak and
brittle with increasing loading of GTR or DGTR [20, 21].
The effect of the waste rubber content on the tensile
performance, so the increase in GTR loading might reduce the strength and the elongation at break were shown in Figs. 7
flow to matrix and consequently increase the torque, resulting and 8. With the amount of waste rubber loading increasing,
in higher viscosity of the compounds. a deterioration of tensile strength and elongation at break of
The maximum torque remained stable with variable GTR NR composites was observed, probably due to the uncon-
loading, while it was lower with an increase of DGTR tinuous and imperfect structure in the blends. It was clear
loading. The torque with DGTR compounds could be that NR/DGTR composites had much better mechanical
explained that partial rupture of cross-linked structure and properties than that of NR/GTR composites at the same
the formation of smaller chains on DGTR surface due to loading, because of the treatment of waste rubber. Although
desulfurization process act as a plasticizer and allow the devulcanization reaction only occurred on the surface of
rubber to regain mobility for better reprocess ability, de- GTR, better compatibility between the matrix and DGTR
creasing the viscosity and torque. led to better properties of NR/DGTR composites. Up to a 10
loading of DGTR, the composites retained 91.7 % of its
Crosslink density original strength and 92.7 % of its elongation at break.
The effect of the waste rubber content on the tear strength
Crosslink density of NR vulcanizates filled with various was shown in Fig. 9. An improvement in tear resistance was
waste rubber content were shown in Fig. 6. The crosslink obtained by the addition of waste rubber to the NR matrix,
density decreased with the waste rubber loading increasing, when waste rubber content was no more than 30 phr loading.
An interpenetrating network structure might exist in such
blends after revulcanization. The tear strength of compounds

Fig.6 Cross-link density vs. waste rubber loading in NR matrix Fig. 7 Tensile strength vs. waste rubber loading in NR matrix
Page 6 of 7 J Polym Res (2012) 19:9864

Fig. 8 Elongation at break vs. waste rubber loading in NR matrix

containing DGTR was much higher than that of the com- Fig. 10 Temperature dependence of loss factor of NR composites
filled with waste rubber at 40 loading content
pounds with GRP because of more active sites existing on
the surface of DGTR.
SEM observation

Dynamic mechanical properties SEM photographs of tensile fracture surfaces of NR com-

posites filled with 30 phr GTR and DGTR were showed in
Figure 10 showed the temperature dependence of loss factor Fig. 11. Figure 11 (a) showed that particle had been pulled
(tanδ) of NR composites filled with GTR or DGTR at a
loading of 40 phr. A remarkable decreased in tanδ peak value
and area of NR composites filled with waste rubber could be
observed, indicating that the addition of waste rubber reduces
the internal friction losses of molecular chains in the glass
transition zone. Compared with the NR/GTR composites, the
tanδ peak value and area of NR vulcanizates filled with DGTR
had a further decrease, suggesting that the surface of DGTR
was more chemically reactive and had better adhesion with the
NR matrix.

Fig. 11 SEM photographs of tensile fracture surfaces of a NR/GTR,

Fig.9 Tear strength vs. waste rubber loading in NR matrix and b NR/DGTR
J Polym Res (2012) 19:9864 Page 7 of 7

out and a clear gap could be observed between GTR and the 4. Hassan MM, Mahmoud GA, El-Nahas HH (2007) Reinforced
material from reclaimed rubber/natural rubber, using electron
matrix. The result indicated that the adhesion of GTR to the beam and thermal treatment. J Appl Polym Sci 104:2569–
NR matrix was poor due to crosslink structure of GTR. It 2578
could be seen from Fig. 11 (b), there was no void between 5. Debapriya D, De D, Singharoy GM (2007) Reclaiming of
DGTR and the matrix. NR/DGTR composites had much ground rubber tire by a novel reclaiming agent. I. Virgin
natural rubber/reclaimed GRT vulcanizates. Polym Eng Sci 47:1091–
better compatibility and phase separation than that of NR/ 1100
GTR. It was agreement with the revulcanization and further 6. Rajan VV, Dierkes WK, Joseph R (2007) Effect of diphenyldisul-
crosslink reaction between DGTR and NR matrix due to fides with different substituents on the reclamation of NR based
much more active sites on the surface of DGTR after desul- latex products. J Appl Polym Sci 104:3562–3580
7. Zhang XX, Lu CH, Liang M (2009) Properties of natural rubber
furization. So the interface region between DGTR and the vulcanizates containing mechanochemically devulcanized ground
matrix could withstand a higher strength and deformation. tire rubber. J Polym Res 16:411–419
8. Du ML, Guo BC, Jia DM (2005) Effects of thermal and UV-
induced grafting of bismaleimide on mechanical performance
of reclaimed rubber/natural rubber blends. J Polym Res 12:473–
Conclusion 482
9. Zhu J, Zhang XX, Liang M, Lu CH (2011) Enhancement of
The surface desulfurization of GTR had been realized with processability and foamability of ground tire rubber powder and
biological treatment by Thiobacillus sp., for improving interfa- LDPE blends through solid state shear milling. J Polym Res
18:533–539
cial adhesion. After treatment, oxygen content on the surface of 10. Balasubramanian M (2009) Cure modeling and mechanical
waste rubber increased by 30 % and cross-linked sulfur bonds properties of counter rotating twin screw extruder devulcan-
were partly cleaved and converted to sulfate or oxygen- ized ground rubber tire-natural rubber blends. J Polym Res 16:133–
containing sulfur-based groups. The ration of S-S bonds and 141
11. Bredberg K, Persson J, Christiansson M, Stenberg B, Holst O
S-C bonds were respectively decreased by 18.3 % and 42.3 %. (2001) Anaerobic desulfurization of ground rubber with the ther-
The formation of S-O bonds was observed. The contact angle mophilic archaeon Pyrococcus furiosus-a new method for rubber
of GTR was 120.5° and decreased down to 93.5° after treat- recycling. Appl Microbiol Biot 55:43–48
ment. The lower crosslink density was obtained for NR/DGTR 12. Kanagawa T, Mikami E (1989) Removal of methanethiol, dimethyl
sulfide, dimethyl disulfide, and hydrogen sulfide from contaminated
vulcanizates and that was attributed to the migration of sulfur air by Thiobacillus thioparus TK-m. Appl Environ Microbiol
being obstructed by GTR with crosslink structure and lower 55:555–558
sulfur content of DGTR. The improvement in mechanical 13. Romine RA, Romine MF (1998) Rubbercycle: a bioprocess for
properties was observed for NR/DGTR vulcanizates because surface modification of waste tyre rubber. Polym Degrad Stab
59:353–358
of the enhanced interfacial interaction between DGTR and NR 14. Sato S, Honda Y, Kuwahara M, Kishimoto H, Yagi N, Muraoka K,
matrix. The results measured by DMA and SEM studies further Watanabe T (2004) Microbial scission of sulfide linkages in
indicated that NR/DGTR composites had lower internal fric- vulcanized natural rubber by a white rot basidiomycete, ceriporiopsis
tion losses of molecular chain and better interface coherence subvermispora. Biomacromolecules 5:511–515
15. Torma AE, Raghavan D (1990) Biodesulfurization of rubber materials.
and homogeneity between the DGTR and matrix. ‘American Society of Mechanical Engineers’ winter annual meeting,
Dallas, pp 81–87
Acknowledgements The project was supported by National High 16. Christiansson M, Stenberg B, Wallenberg LR, Holst O (1998)
Technology Research and Development Program of China (No. Reduction of surface sulphur upon microbial devulcanization of
2006AA06Z367), National Nature Science Foundation of China (No. rubber materials. Biotechnology letters 20:637–642
50673009), and Nature Science Foundation of Beijing (No. 8122032). 17. Flory PJ, Rehner JJR (1943) Statistical mechanics of cross-linked
polymer networks II. Swelling. J Chem Phys 11:521–526
18. Mathew G, Singh RP, Lakshminarayanan R, Thomas S (1996) Use
References of natural rubber prophylactics waste as a potential filler in
styrene-butadiene rubber compounds. J Appl Polym Sci 61:2035–
2050
1. Adhikari B, De D, Maiti S (2000) Reclamation and recycling of 19. Moulder JF, Stickle WF, Sobol PE, Bomben KD (1992) Handbook
waste rubber. Prog Polym Sci 25:909–948 of X-ray Photoelectron Spectroscopy. Perkin Elmer Corporation,
2. Fukumori K, Matsushita M, Mouri M (2006) Dynamic devulcani- Minnesota
zation and dynamic vulcanization for recycling of crosslinked 20. Ismail H, Nordin R, Noor AM (2002) Cure characteristics, tensile
rubber. Kaut Gummi Kunst 59:405–411 properties and swelling behaviour of recycled rubber powder-filled
3. Vega B, Montero L, Lincoln S (2008) Control of vulcanizing/ natural rubber compounds. Polym Test 21:565–569
devulcanizing behavior of diphenyl disulfide with microwaves as 21. Phadke AA, Bhowmick AK, De SK (1986) Effect of cryoground
the heating source. J Appl Polym Sci 108:1969–1975 rubber on properties of NR. J Appl Polym Sci 32:4063–4074