Manuscript12 1 23
Manuscript12 1 23
Manuscript12 1 23
Abstract
The current focus in the elastomer industry is to look for biobased or natural filler for various
compounds. The reinforcing effect of waste drumstick-derived cellulose micro and nanofiber in
styrene butadiene rubber (SBR) was investigated for the very first time. Cellulose microfibers and
nanofibers were isolated from the drumstick fruit peel and characterized using different techniques
such as X-ray diffraction, scanning and transmission electron microscopy, and X-ray photoelectron
spectroscopy. The filler-polymer interaction in the composites was studied using the bound rubber
measurements which showed that the bound rubber content of the composite increased from 0.5%
to 8.1% at 5 phr cellulose nanofiber loading. The tensile strength and modulus of the modified
cellulose microfiber-containing composite were higher than both the unfilled SBR and the raw
fiber-SBR composites. The addition of 5 phr nanofiber to SBR showed a 46% increment of the
tensile strength than the control compound. The die swell index of the cellulose fiber reinforced
composites came down attesting the better processability of the extrudates. The addition of
cellulose fiber also improved the wet skid resistance of the composites as seen from dynamic
mechanical thermal analysis. Hence, this work confirms the application of modified drumstick
1
1. INTRODUCTION
Rubber is a soft material with high elongation at break. The incorporation of reinforc ing
materials in the soft rubbery domain improves the mechanical properties of the elastomer. Carbon
black is the primarily used reinforcing material for rubber products. However, the non-
biodegradable nature and environmental hazards of carbon black motivate researchers to work
The requirement for high strength and flexibility with lightweight has escalated the market
importance of fiber in the field of composites. As the most abundant biopolymer and the main
constituent of plant fibers, cellulose (C6 H10 O5 )n is of prime importance. Use of the natural
cellulosic fiber also provides a sustainable solution to the disposal problem with an additional price
and energy benefit. Natural fibers can be classified as bust fiber, seed hairs, leaf fiber, grass fiber,
and fruit peel fiber.1 Properties of the fiber also depend on the source, fiber extraction process, and
aspect ratio of the fiber. Effects of natural fibers like silk short fiber, jute, sisal, bagasse, and palm
on polymer composite have also been investigated.2-5 Alkali-treated cellulose short fiber reinforced
natural rubber (NR) composite showed improved crosslink density with higher tensile strength.6
mechanical size reduction are also investigated by different researchers.3,7 Depending on the
diameter of the extracted fiber, it can be classified as microfiber (~ 44 μm) or nanofiber (˂100
nm).8 At higher loading, high moisture affinity, low thermal stability, and poor compatibility with
nonpolar rubbers restrict the use of cellulose microfibers in rubber composites. On the other hand,
the high surface area of cellulose nanofiber can improve the properties of the compound even at a
very low filler loading.9 Different fiber surface modification methods like silane treatment, stearic
2
acid treatment, 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO) mediated oxidation, etc., are
Among different natural sources of fiber, the drumstick is also a potential source of cellulose.
Moringa oleifera is a tropical tree abundant in India, Ethiopia, Asia, Florida, etc.12,13 The hard
outer layer of the drumstick, which is thrown away like waste, can be used as a potential source of
cellulose micro and nanofibers. The use of renewable resource-based materials and waste
management are the two important pillars of sustainable development. Bharath et al. have recently
reported the effect of alkali treatment on the properties of drumstick fiber (DF).14 The general
properties of drumstick fiber like light weight, high specific modulus, moisture adsorption
tendency, and non-toxic nature are similar to any other untreated cellulose fiber. But the average
density of the alkali-treated moringa fiber (1.14 g/cm3 ) is comparable to the density of hemp fiber
(1.14 g/cm3 ).15 As per the literature, the elongation at break of the untreated drumstick fiber (>
4%) is higher than the breaking elongation of hemp fiber (1.6 %).15 The reinforcing effect of the
alkali-treated drumstick fiber, drumstick leaves-derived fillers, and drumstick nanofiber on the
properties of the epoxy resin composite was studied by different research groups.16-18 Drumstick
fiber was also utilized with red mud for the reinforcement of chitosan.19 . Numerous investigatio ns
are available on cellulose nanofiber reinforcement of natural rubber composite. But very few
et al. studied the effect of drumstick nanofiber on the tensile and thermal stability of the NR-latex-
based film.20 However, no study is available on the surface modification of the drumstick peel-
derived fibers using different surface modification methods and their effect on synthetic rubber. A
detailed study of the microscale and nanoscale drumstick fiber reinforcement on the mechanica l
3
Present work was conducted to explore the possibility of using drumstick–derived cellulose
micro and nanofibers to reinforce rubber composites. This work also presented the correlation
between differently modified fiber properties and their reinforcing effect in styrene-butadie ne
rubber. The working hypothesis was the extracted and modified cellulose microfiber and
nanofibers of different compositions and polarity would impart improved reinforcement to the
fiber-elastomer composites. Water retting followed by alkali treatment released the microfibers
from the drumstick peel. The hydrophobicity of the microfiber was enhanced by stearic acid and
containing SBR composites with both the unfilled and raw fiber-containing compounds is
presented in the work. Cellulose nanofibers, produced by acid hydrolysis of the raw drumstick
microfiber, were also mixed with SBR. Properties like the mechanical, curing, and processability
of the cellulose nanofiber- filled compound were compared with the unfilled SBR-based
compound.
2.1. Materials
Drumstick fiber was collected from local sources. SBR1502 (ML1+4 at 100 °C = 51) was
purchased from Kumho Petrochemical, South Korea. Sodium chlorite (80% extra pure) and sulfur
were bought from Loba Chemie Pvt. Ltd., India. Bis-(3-triethoxysilylpropyl) tetrasulfide (TESPT)
was purchased from Momentive, India. ZnO was procured from Merck, India. Stearic acid was
(6PPD) and cyclohexyl benzothiazole sulfenamide (CBS) were supplied by National Organic
4
2.2. Methods
Initially, the drumstick peel was adequately washed to remove the impurities and soaked in
water for 1 week at room temperature. Expulsion of the upper and bottom layer released the
cellulose fiber which was dried at 70˚C for 24h. The oven-dried fibers were chopped into fibers of
3-4 mm in length. The yield of the drumstick-isolated raw cellulose fiber was 46.6%.
2.2.2. Alkali treatment of the raw fiber for the isolation of microfiber (AF)
The chopped RF was digested with 4 wt % NaOH solution at 80 ˚C for 4h. The isolated
cellulose microfiber was dried at 70 ˚C for 24h. The obtained yield of the microfiber was 64.5%
ethanol-water mixture and heated at 80 ˚C for 2h (Scheme S.Ⅰ, in supporting information). The
fiber was dried at 70 ˚C for 24h. The yield of the stearic acid-treated fiber was 91.4% of the used
pH 3.5 to 4 and kept undisturbed for 1h. Then alkali-isolated microfibers were soaked in the
solution for 2h at room temperature. The fibers were thoroughly washed with ethanol and dried at
70 ˚C for 24 h, followed by heat treatment at 110 ˚C for 8h (Scheme S.Ⅱ, in supporting informatio n)
and the obtained modified microfiber yield was 94.3% of the used alkali-isolated microfiber of the
drumstick.
5
2.2.5. Extraction of cellulose nanofibers (NF) from waste drumstick peel
Raw drumstick fibers were placed in a 2:1 benzene/ethanol mixture in a soxhlet apparatus.
The extraction was continued at 60 ˚C for 7h. Fibers were thoroughly washed with distilled water
and soaked in DMSO, followed by treatment with 4 wt % NaOH at 80 ˚C for 4h. Then the alkali-
treated drumstick fiber was bleached with 2 wt % CH3 COOH acidified solution of sodium chlorite
(pH 2.5) at 75 ˚C for 1h. This was repeated thrice and washed with water to get a neutral pH. The
bleached fiber was hydrolyzed using 40 % sulphuric acid at 45 ˚C for 130 min.20,21 The reaction
was terminated by adding an excess of cold distilled water. The cellulose nanofiber was collected
by centrifugation for 15 min at 10000 rpm followed by dialysis in water. The obtained nanofibers
Compounding of the SBR with raw drumstick fiber, drumstick-derived micro, and nanofibers
was done as per the standard formulation presented in Table S.1 (in supporting information). The
mixing of SBR with compounding ingredients and fibers was conducted in a Haake Rheocord
internal mixer at 100 ˚C and 60 rpm. First, the rubber was masticated for 2 min followed by the
addition of ZnO, stearic acid, and 6PPD. At 3rd minute , fibers were added to the Hakke Rheocord
mixer and sheared for 7 min After the maturation period, the accelerator was mixed with the master
batch for 1.5 min followed by mixing of sulfur for another 1.5 min in a laboratory-size two-roll
mill.
2.3. Characterization
Fourier transform infrared spectra (FTIR) of the fibers were studied on a Perkin-Elmer
spectrophotometer. The scanning range was fixed to 4000-400 cm-1 with a resolution of 4 cm-1 .
6
X-ray photoelectron spectroscopy (XPS) studies of the fiber were conducted in an Omicron
Wide-angle X-ray spectra (XRD) of the fibers were recorded in a Bruker D-8 advance Eco X-
ray powder diffractometer with monochromatic Cu-Kα radiation of 0.15404 nm. The crystallinity
percentage and crystal thickness of the fiber were calculated using Equation (S.Ⅰ) and Equation
Dimension and the dispersion of the fibers in the rubber matrix were studied by Carl Zeiss
MERLIN scanning electron microscope (SEM) attached with a field emission gun of 5 kV
accelerating voltage. Gold-coated cryo-fractured surface of the composite was chosen for the
experiment.
The dimension of the extracted cellulose nanofibers was also confirmed by using an analytica l
transmission electron microscopy (TEM) with a FEI Tecnai electron microscope operating at 120
fibers.
Thermal transitions of the fibers were studied using a NETZSCH DSC 200 F3 Maia differentia l
The effect of the drumstick fibers on the die swell and processability of the compounds was
studied in a Monsanto capillary rheometer with a 20:1 L/D ratio. The extrusion was conducted at
The Cure characteristics of the rubber-fiber composites were measured with a rubber process
7
A rubber sheet of 1 mm thickness was molded in an electrically heated press at 150 ˚C
temperature and 5 MPa pressure for the optimum cure time. After conditioning, dumbbell-shaped
The tensile test was carried out as per the ASTM D 412 using a Universal Testing Machine
(Hounsfield H25KS) at room temperature. The test speed was fixed at 500 mm/min. The tear
strength and Shore A hardness of the composites were determined by following ASTM D 624 and
ASTM D 2240, respectively. The statistical significance of the obtained results was determined by
For the crosslink density measurement, disk-shaped molded samples were immersed in toluene
at room temperature. The test was continued until equilibrium was reached. The crosslink density
of the rubber was calculated using the Flory-Rehner equation, given as Equation (S.Ⅲ) (in
supporting information).
For bound rubber content (BRC) measurement, the unbound part of the specimen was extracted
by using toluene as a solvent for 7 days at room temperature. The bound rubber content was
instrument under a nitrogen atmosphere. Approximately 10 mg of the compound was taken for the
test. The temperature was increased from room temperature to 800 ˚C at a 10 ˚C /min heating rate.
Dynamic mechanical analysis (DMA) of the composites was carried out using a DMA Q800
instrument supplied by TA Instruments at 1 Hz frequency and 0.1% constant strain. The sample
of 35 mm in length and 6 mm in width was punched from the previously molded sheet. For the
temperature sweep, the temperature was varied from −70 to 80 ˚C at a heating rate of 10 ˚C /min.
8
3. RESULTS AND DISCUSSION
The main components of the drumstick fiber are cellulose, hemicellulose, and lignin. The
primary functional groups of the constituent materials are hydroxyls, alkanes, aromatics, esters,
and ketones. The comparative FTIR spectrum is presented in Figure 1. For all the fibers, the
obtained peak at 3333 cm-1 represents the stretching vibration of the present –OH groups in the
main cellulose structure, whereas the bending mode vibration of the absorbed water showed a
signal at 1648 cm-1 .22 For the water-retted raw drumstick fiber, the ester carbonyl group stretching
was observed at 1732 cm-1 . This was for the acetal, uronic acid ester of the hemicellulose, and p-
coumaric acid, and ferulic acid esters of lignin present in the fiber after water retting. The aromatic
C=C stretching of the lignin structure and out-of-plane stretching of C-O-C ether linkage present
in raw fiber displayed signals at 1504 cm-1 and 1241 cm-1, respectively.3,22 The broad peak at 1030
Alkali digestion endorsed the cleavage of present ester bonds and expedited the liquefactio n
of the remaining hemicellulose and the lignin part of the water-retted drumstick fiber. The
disappearance of the 1732 cm-1 peak in FTIR of the alkali- isolated microfiber (AF) confirmed the
previous statement. Both the stearic acid-treated and TESPT-modified cellulose microfiber
showed an increase in the intensity of the 2898 cm-1 peak than the AF. The long hydrocarbon chain
of stearic acid and TESPT present on the surface of modified cellulose microfiber might be the
reason for the increased intensity of the 2898 cm-1 peak. The increased intensity of the 1107 cm-1
peak for Si-O-C linkage confirmed the effective TESPT modification of the cellulose microfibers.
For the synthesized cellulose nanofiber, the 1732 cm-1 , 1504 cm-1 , and 1241 cm-1 peaks for
lignin were absent in the FTIR spectrum. Alkali treatment of the raw drumstick fiber released most
9
of the lignin and hemicellulose present in the fiber. During the bleaching stage of the nanofiber
synthesis, acidified sodium chlorite released chlorine oxide, which further reacted with the
remaining phenolic components of the lignin even after alkali treatment. It freed the cellulose
The surface elemental composition of the raw drumstick fiber, isolated microfiber, and
surface-modified microfibers was investigated by XPS. The obtained spectrum of the raw
drumstick microfiber is presented in Figure 2. The XPS spectra of the AF, SF, SiF, and NF are
shown in Figure S.1a to Figure S.1c and Figure S.1d, respectively (in supporting information). All
the fibers showed a typical spectrum for carbon around 285.0 eV and an O signal at 533 eV.
10
However, a significant change in C and O ratio was observed for the cellulose micro and
nanofibers, as presented in Table 1. The XPS signal for carbon atoms combines the contributio n
of all the carbon-containing compounds present in the fiber. Each compound's contribution was
determined using the deconvolution method and reported in Table 1. The O/C ratio for the raw
drumstick fiber was 0.54, which was lower than the theoretical O/C ratio (0.83) of pure cellulose. 2 4
The lower O/C ratio of the raw drumstick fiber (0.54) than the pure cellulose suggests the presence
of lignin (0.35) and wax with the major cellulose component of the fiber. Deconvolution of the
C1S spectrum of raw drumstick fiber also revealed the presence of different carbons like cellulose,
lignin, hemicellulose, and wax. Peaks at 284.8, 286.1, and 287.5 eV confirmed the presence of C-
C, C-O, and O-C-O/C=O linkages in the raw fiber, respectively. The FTIR peak at 1732 cm-1 , and
the XPS signal at 287.9 eV were for the O-C=O linkage of the ester groups present in the RF.
Another signal at 283.1 eV was for the difference in neutralization time of the loosely bounded
fibrils than the bulk fiber.25 Similarly, deconvolution of the high-resolution O1S spectrum of the
raw fiber also produced two signals for O-H (531.3 eV) and O-C (533.9 eV) linkages.
Alkali treatment promoted the dissolution of lignin, hemicellulose, and other carbonaceous
parts of the fiber. Hence, the O/C ratio of the alkali-isolated microfiber increased from 0.54 to
0.70.
Stearic acid and silane treatment increased the unoxidized carbon concentration at the fiber's
surface. Hence, both SF and SiF showed a hike in C-C bond percentage from 35.4 % (for alkali-
treated fiber) to 57.2% and 53.5%, respectively. The appearance of a new ester peak at 288 eV
confirmed the stearic acid modification of the drumstick microfiber. Silane-modified microfiber
showed clear signals at 101.5 eV and 151.5 eV for Si2P and Si2S, respectively. An increase in silicon
11
percentage from 1.8 (AF) to 5.9 % suggested the effective silane modification of the microfibers.
For the synthesized cellulose nanofiber, a series of chemical treatments like alkali treatment
followed by bleaching and sulphuric acid treatment removed the lignin and hemicellulose part of
the raw fiber. Allied with the FTIR spectra of the synthesized cellulose nanofiber, the XPS
spectrum also showed the absence of any signal for the ester or carbonyl linkage of lignin and
hemicellulose. For synthesized cellulose nanofiber, the O/C ratio was 0.66, which is very close to
the reported value for cellulose nanocrystal (0.69).26 The loosely bounded carbon percentage is
also meager for the drumstick-derived cellulose nanofiber. High crystallinity with a tightly
hydrogen-bonded structure might be the reason for the decrease of loosely bonded carbon of the
12
TABLE 1. Elemental composition and percentage of different linkages present in the fibers.
Fiber O/C Si2p Signal for loosely C-C C-O O-C-O/ O-C=O O-H C-O
% bounded carbon % % C=O % % %
% %
RF 0.54 1.7 21.4 37.7 15.6 12.8 12.5 62.4 37.6
SEM analysis was conducted to get an intelligible idea about the effect of different chemical
treatments on the diameter and surface features of the drumstick-isolated cellulose micro and
nanofiber. SEM image of the segregated raw drumstick fiber is presented in Figure 3.a. and the
average diameter of the fiber was 81±13 μm. SEM image of the raw fiber surface (in Figure 3.b)
clearly showed the embedded cellulose fiber in the lignin matrix. The diameter of the drumstick-
separated cellulose microfiber was 26 ±10 μm, as calculated from Figure 3.c. Dissolution of
present lignin, and hemicellulose during alkali treatment released the fiber of lesser diameter. A
similar observation was also reported by Mwaikambo et al. The diameter of the raw and alkali-
treated hemp fiber was reported as 67.84 μm and 38.35 μm, respectively.27 Alkali treatment also
increased the surface roughness of the microfiber, which is visible in Figure 3.d. Stearic acid
modification further helped to dissolve the remaining lignin, and defibrillation occurred (Figure
3.e-f). The average diameter of the stearic acid-treated microfiber was 23±7 μm. The diameter of
the TESPT-treated microfiber (27±13 μm) was comparable to the alkali- isolated microfiber
diameter (Figure 3.g). The surface appearance of the SiF is presented in Figure 3.h.
13
In the case of cellulose nanofiber synthesis, alkali treatment and bleaching stage removed
most of the lignin percentage from the raw microfiber. The remaining amorphous part was also
very prone to sulphuric acid attack during the nanofiber synthesis. Hence, sulphuric acid-generated
hydronium ions can easily penetrate the remaining amorphous region and hydrolyze the glycosid ic
bonds to form the cellulose nanofiber. The average diameter of the synthesized cellulose nanofiber
as calculated from SEM was 24±7 nm (Figure 4.a-b). TEM analysis (shown in Figure 4.c-d) also
confirmed the synthesis of cellulose nanofibers (22±7 nm diameter) from raw drumstick fiber. As
per literature, 64 % sulfuric treatment of the hemp fiber at 45 ℃ for 4 h also produced cellulose
14
FIGURE 3. Diameter and surface appearance of the RF (a. & b.), AF (c. & d.), SF (e. & f.), and
15
FIGURE 4. SEM (a & b) and TEM (c & d) images of the synthesized cellulose nanofiber (NF)
The crystalline structure of the cellulose fibers was explored using the XRD, and the
comparative graph is presented in Figure 5.a. The result showed three characteristic peaks at 2θ of
15.7˚, 22.7˚, and 34.4˚ for the (101), (002), and (004) crystallographic planes, respectively. The
prevailing crystalline peak of all the fibers appeared at 22.7°, and the percentage crystallinity of
all the fibers was calculated based on the 22.7° peak using Equation (S.I) (in supporting
information). Obtained percentage crystallinity of the RF and AF were 34.7 % and 57.8 %,
respectively. Partial liquification of the lignin and hemicellulose during alkali treatment might be
16
During stearic acid treatment, the carboxylic group of the acid reacted with the hydroxyl group
of cellulose and formed the stearate ester. Carbonyl groups on cellulose surface can participate in
hydrogen bonding, and the percentage crystallinity slightly increased to 58.1%. During silane
treatment, partial substitution of the hydroxyl groups with TESPT reduced the hydrogen atom
accessibility for the hydrogen bonding, and the crystallinity marginally decreased from 57.8% to
57.1 %.
For the synthesis of cellulose nanofiber, raw drumstick fiber was pretreated with chemica ls
like alkali and sodium chlorite. Hence, a large amount of the amorphous lignin and hemicellulose
part was dissolved in the pretreatment step. During sulphuric acid hydrolysis, the formed
hydronium ion attacked the glycosidic bond of cellulose and produced cellulose nanofiber from
the pretreated cellulose microfiber. Proper alignment of the formed crystallites increased the
percentage crystallinity of cellulose nanofiber to 62.3 %. XPS result also supported the previous
statement. During chemical modification, accumulated strains between the amorphous and
crystalline domains of the cellulose decreased, leading to increased crystallite thickness. The
calculated crystallite size of the raw drumstick fiber, alkali-isolated microfiber, silane-treated, and
stearic acid-treated microfiber were 2.97, 3.79, 4.10, and 4.20 nm, respectively. Previous literature
also reported a similar trend.29 The Crystallite size of the cellulose nanofiber was 4.18 nm.
17
FIGURE 5.a. XRD pattern and b. DSC of the drumstick-derived cellulose microfibers and
nanofibers
DSC analysis of the fibers gives an idea about the effect of chemical treatment on the thermal
transition of the cellulose fibers. A comparative DSC thermograph of the fibers is presented in
Figure 5.b. For all the fibers two distinct endothermic peaks were obtained within the studied
temperature range. The first endotherm below 150 ˚C was for the evaporation of the moisture
absorbed by the cellulose fibers. RF peaked at 126 ˚C, whereas all the modified and unmodified
microfibers showed the endotherm at around 94 ˚C. A large amount of hydrophilic lignin and
hemicellulose content was responsible for the higher moisture evaporation temperature of RF.
Different chemical modifications removed the hydrophilic hemicellulose and lignin part leading
Sulphuric acid hydrolyzed cellulose nanofibers have a sulfate functionalized surface, which
exceptionally lessens the moisture affinity of the cellulose. Therefore, moisture evaporation
18
Fusion of the crystalline part of both the unmodified and the modified microfibers showed a
second endothermic peak around 361 ˚C. The lower molecular weight of the cellulose nanofiber
shifted the cellulose fusion peak to a lower temperature (339 ˚C). Thermal degradation of the
glycosidic linkages of the cellulose and hemicellulose structure of raw fiber gave another exotherm
at around 291 ˚C.31 The absence of such a pyrolysis peak confirmed the purity of the synthesized
cellulose nanofiber. Charring of the raw drumstick cellulose also registered two exothermic peaks
at 378 ˚C and 428 ˚C.32 The exothermic transition for the charring of cellulose nanofiber appeared
at 363 ˚C. The lower degree of polymerization and acid degradation might be the reason for the
The effect of the fibers on the shear viscosity, shear stress, and die swell of the composites are
presented in Figure 6. With the addition of fibers, no significant change was observed in the initia l
viscosity of the compounds compared to the unfilled SBR-based compound (Figure 6.a). But, at a
500 s-1 shear rate, the shear stress (Figure 6.b) of all the cellulose microfiber-containing
compounds was slightly higher than the control compound. This is due to the hydrodynamic effect
At a higher shear rate, the viscosity of the cellulose nanofiber-containing compound decreased
below the control compound viscosity (Figure 6.a). Progressive disruption of the cellulose
nanofiber network is responsible for reducing the compound viscosity at a high shear rate. The
shear stress of the cellulose nanofiber-containing compound at a high shear rate was also
The die swell for the entire fiber-SBR composites (Figure 6.c) was lower than the control
compound. During extrusion, rubber chains get aligned in the direction of extrusion. But
19
elastomers have a natural tendency to recoil and, the die swell occurs. Added cellulose fibers
adsorbed the rubber on the filler surfaces and restricted the recoiling of the elastomers. Hence, the
die swell property of the SBR-fiber composites was better than the control compound.
FIGURE 6. Effect of different cellulose fibers on a. Viscosity, b. Shear stress and c. Die
20
3.7. Effect of the cellulose micro and nanofiber on curing of the compounds
The effect of different fibers on the cure properties of the SBR was studied, and the rheograph
is presented in Figure S.2 (in supporting information). Cure properties of the compounds like
optimum cure time (tc90 ), maximum torque (MH), and minimum torque (ML) are shown in Table
2. The addition of any cellulose fiber increased the MH of the compounds, and the values further
improved with the increase of the fiber loading of the compounds. The inclusion of the fibers
restricted the mobility of the macromolecular chains leading to the rise in the viscosity and stiffness
of the composites. Enhanced surface roughness is the reason for the higher rubber filler interactio n
and maximum MH of the alkali- isolated microfiber-containing materials. The MH values of all the
The cure rate index of the RF-containing compound was slightly higher than the control
compound. Present natural fatty acids and waxes acted as activators for curing the composite. For
both the RF and AF-containing compounds, scorch safety decreased. This can be the result of
increased mixing time and generated heat while mixing the fibers in rubber.33 Further defibrilla tio n
and increased acidic nature of the stearic-treated microfiber increased the cure and scorch time of
the composite.
The tc90 and ts2 of the cellulose nanofiber-SBR mix were higher than the control compound.
The acidic nature of the cellulose nanofiber might be the reason for increased optimum cure time
and a decrease in the cure rate index of the composite.34 Due to the large active surface area,
cellulose nanofiber can adsorb the curatives, and tc90 increased. Adsorption of the ZnO activator
on the surface of cellulose nanofiber by forming a hydroxyl group-metal oxide complex also
increased the scorch safety of the nanocomposite than the unfilled control compound.35
21
TABLE 2. Cure and mechanical properties of all the fiber-reinforced composites
Properties SBRC SBR5RF SBR10RF SBR5AF SBR10AF SBR5SF SBR5SiF SBR3NF SBR5NF SBR7NF
Maximum torque 9.13 10.22 11.36 10.81 12.62 10.7 10.63 9.68 10.1 10.25
(MH), dN.m
Minimum torque 0.62 0.71 0.81 0.75 0.87 0.75 0.76 0.53 0.71 0.73
(ML), dN.m
Scorch time (ts2 ), 7.52 6.57 6.27 7.17 7.40 8.47 8.30 8.78 8.36 8.92
min
Optimum cure 15.59 13.99 13.62 15.54 15.76 17.06 16.43 17.82 16.78 17.25
time (tc90 ), min
Cure rate index 12.39 13.48 13.61 11.95 11.96 11.64 12.30 11.06 11.87 12.00
(CRI), min-1
Tensile strength, 1.3±0.1 1.0±0.1* 1.0±0.1* 1.7±0.0* 2.3±0.2* 1.8±0.0* 1.7±0.1* 1.7±0.0* 1.9±0.0* 1.8±0.0*
MPa
100% Modulus, 0.8±0.0 - - 1.6±0.1* - 1.6±0.0* 1.5±0.1* 1.0±0.1* 1.2±0.0* 1.3±0.0*
MPa
EL@B, % 232±22 88±6* 46±11* 140±14* 56±8* 199±13* 157±2* 271±13* 301±4* 268±11*
Tear strength, 11.1±0.2 9.4±1.2* 13.2±0.8* 13.6±0.3* 14.7±2.0* 13.0±0.7* 12.8±0.9* 13.2±1.7* 13.4±0.9* 14.7±1.2*
MPa
Hardness, shore 53±1 60±1* 65±1* 63±1* 68±1* 62±2* 61±1* 54±1* 54±1* 57±1*
A
Crosslink 3.12 3.29 3.11 3.31 3.28 3.39 3.38 3.32 3.52 3.39
density, mol/c.c.,
×10-4
*Asterisk indicates that the difference in the mean is statistically significant at the level of 0.05
compounds
The effects of different fibers on the mechanical properties of the compounds are reported in
Table 2. The addition of RF reduced the tensile strength and elongation at the break of the
compound than the unfilled control compound. Lignin and other impurities of the RF preclude the
bonding of the fibers with the rubber matrix. Dewetting of the raw microfiber surfaces by the SBR
matrix, and the agglomeration of the cellulose fibers create stress concentration points and reduce
22
the composite's tensile strength.33 Kumar et al. also reported a decrease of tensile strength with the
addition of 5 phr of the short sisal fiber in a SBR-based composite.36 Similar effect was also
reported for the ground pistachio shell containing NR/SBR-rubber-based compounds, and the
value was decreased from 8.24 to 6.99 MPa.37 Raza and coworkers studied the decrease of tensile
strength of the 5 phr raw eucalyptus kraft pulp containing SBR composite than the unfilled control
compound.38 As per Rueda et al., 30 phr cellulose short fiber loading reduced the tensile strength
of a sepiolite filled (35 phr) SBR composite from 4.9 to 4.36 MPa.39 Haghighat et al. reported
approximately a 16.7 % rise of tensile strength with the addition of 5 phr of cotton-based α-
cellulose of 150 µm particle size to a 30 phr carbon black filled SBR composite.40 The addition of
both the raw drumstick fiber and the isolated microfibers reduced the matrix's elasticity by
restricting the recoiling of rubber chains, and the bound rubber content of the composites increased
(Figure 7). Improved rubber-filler interaction also increased the tensile strength of the SBR5AF.
The alkali-treated drumstick fiber (microfiber) at 10 phr loading, increased the tensile strength of
the SBR-based compound by 76.9 % than the control compound. Depending on the mechanica l
properties like tensile strength and elongation at break, 5 phr fibers loading was considered
optimum. All the modified microfibers were mixed at 5 phr loading, and the comparative
mechanical properties are presented in Table 2. The increment of tensile strength for the 5 phr
stearic acid and silane-treated drumstick fiber-containing compounds were 38.5 % and 30.8 %
respectively. Raza et al. also report the marginal improvement of the tensile strength for the 5 phr
maleated high oleic sunflower oil (MSOHO) treated fiber reinforced SBR compound.38
Reduction in polarity and decreased diameter of SF enhanced the SBR-SF interaction leading
to the increase of tensile strength and elongation of the composite than both the RF and AF-
23
containing compounds.33 Modulus of both the SBR5SF and SBR5SiF was higher than the SBRC
and SBR5AF.
The addition of NF at 3, 5, and 7 phr loading, increased the tensile strength of the composites
compared to the unfilled control compound by 31%, 46%, and 38%, respectively. At 5 phr loading
was reported for the SBR-based composite.38 Hence, the obtained improvement of tensile strength
values in the literature. The reinforcing efficiency of the cellulose nanofiber (CNF) depends on the
sources, processes of fiber extraction from natural sources, and the compounding. As per the
literature, solution mixing of cellulose nanofiber with unfilled SBR increased the tensile strength
of the composite from 3.2±0.71 to 8.06±0.95 MPa.41 Many works are available on cellulose fiber
reinforcement of natural rubber.42,43 Kazemi et al. reported an increase in tensile strength from
14.7 MPa to 18.7 MPa (27.2 % improvement) by adding 5 phr of nanocellulose in the melt mixing
process.44 Jiang and coworkers reported around 60 % improvement of the tensile strength of NR
composite at 5 phr nanocellulose mixing in the latex phase.45 Kanoth et al. showed a 50 %
improvement of the tensile strength (from 2.4±0.4 MPa to 3.6±0.4 MPa) at 5 phr loading of the
Uniformly dispersed nanofillers form a fiber network in the rubber matrix and can evenly
transfer the stress throughout the matrix. Hence, the tensile strength of the composite enhanced.
But at higher loading (7 phr), hydrophilic cellulose fiber started agglomerating and tensile strength
decreased. But, the modulus and hardness of the SBR7NF were higher than the other nanofiber
rubber composites. The incorporation of rigid filler particles increased the modulus of the
24
composites with the reduction of elongation at the break. But, SBR3NF showed an increase in
elongation at break from 232% to 271% than SBRC. Maximum elongation (301%) was obtained
at 5 phr NF loading. During the tensile test, the physical interaction points between the NF and
rubber oriented the nanofibers in the stretching direction of the sample and improved the
elongation at break.47 Further increase of cellulose nanofiber loading again restricted the SBR
chain mobility, and elongation at break decreased. A proper balance between reinforcement and
The tear strength of the compounds gives an idea about the crack growth resistance of the
material under tension. As reported in Table 2, the tear strength of the SBR5RF was lower than
the unfilled SBR-based compound. Poor rubber-filler interaction in the RF-containing compound
created weak points in the composite, and tear strength decreased. But, the removal of surface
impurities like lignin, and hemicellulose increased the surface roughness of AF. Hence, due to the
better rubber-fiber interaction, SBR5AF showed higher tear strength than the SBRC and SBR5RF.
For the NF-containing compounds, the high surface area and uniform distribution of the fibers in
the rubber matrix obstructed the microcracks propagation in the sample, and tear strength
increased.48
The swelling of all the fiber-containing composites was studied and presented in Table 2. The
crosslink density of all the cellulose fiber-SBR composites was higher than the control compound.
The addition of filler formed a tortuous path for solvent diffusion and hindered the swelling of the
rubber.49 It also decreased the free rubber fraction of the composites. Increased bound rubber
content (Figure 7) of all the fiber-filled compounds reinforced the previously mentioned theory of
25
restricted solvent diffusion. The crosslink density of SBR5SF and SBR5SiF was comparable and
SBR5NF showed a prominent rise in crosslink density than the control compound. It is due to
the large surface area and better rubber nanofiller interaction. Interaction of the fiber hydroxyl
group with the added ZnO also helped to hold the rubbers on the fiber surface firmly and acted as
a crosslinking point during swelling. The crosslink density of the cellulose nanofiber-containing
compound further decreased for the SBR7NF. At higher loading, cellulose nanofiber starts to
agglomerate and the crosslink density decreased with the reduction of rubber-filler interaction.
Bound rubber content gives an idea about the extent of interaction between rubber and
cellulose fibers in the composite. As an excellent solvent, toluene could quickly diffuse into the
composite and almost completely dissolve the rubber for the control compound (0.5%). For the
entire microfiber-rubber composites, bound rubber content was higher than the control compound
(Fig. 7). Increased hydrophobicity of SF and SiF enhanced the rubber-fiber interaction, and the
Bound rubber content values of all the cellulose nanofiber-SBR composite were higher than
the control compound. Among all the cellulose nanofiber-SBR composites, maximum bound
rubber content was observed at 5 phr loading of NF. Because of the large effective surface area
with the highly entangled structure of the NF, the trapped rubber amount in the fiber network
decreased. Hence, the bound rubber content of the SBR7NF was lower than the SBR5NF.
26
FIGURE 7. Bound rubber content of the drumstick isolated fiber containing composites.
The effect of the cellulose micro and nanofiber on the thermal stability of the SBR-based
composites was studied using the thermogravimetric analysis and presented in Figure 8. Thermal
degradation of the low molecular weight compounding agent started around 150 °C and gave a
small peak at about 225 °C. The maximum decomposition of SBRC was obtained at 433 °C. The
required temperature for 20% degradation (T20 ), maximum degradation temperature (Tmax ), and
the remaining residual part of the composites at 780 °C are presented in Table S.2 (in supporting
information). As the cellulose fibers are prone to thermal degradation, T20 and Tmax values of the
SBR5RF were lower than the SBRC.12 For the SBR5AF composite, increased rubber-filler
interaction and rising crosslink density hindered the diffusion of volatile decomposition products
from the compound, and the thermal stability of the compound was comparable to SBRC.48 Surface
modification of microfibers and the synthesis of cellulose nanofiber by acid hydrolysis further
27
reduced the diameter of the fibers leading to a decrease in the Tmax and T20 values of the composites
Figure 9a, 9b, and 9c respectively present the effect of different fillers on storage modulus,
loss modulus, and glass transition temperature of the SBR at 5 phr loading. Figure 9a showed the
increase of storage modulus by adding the fibers to SBR. The fibers restricted the movement of
the polymer chains leading to a rise in the storage modulus of the composites. The storage modulus
curve of SBRC, which was only for the unfilled SBR, showed a sharp drop at -36.9 °C. The chain
relaxation during the transition from the glassy to the rubbery zone is the reason for the sharp
decrease of the storage modulus of the compound. Below the glass transition temperature, the
effect of the fiber reinforcement is more prominent than the storage modulus at 25 °C. Improved
rubber fiber interaction enhanced the storage modulus of the SBR5SiF compared to other
microfiber-SBR compounds.
28
In case of the SBR-cellulose nanofiber composites, nanofibers formed an interconnected
network at a concentration higher than the percolation threshold and increased the adsorbed rubber
content of the composite.50 The storage modulus of the NF-containing SBR compound also
increased.
The loss modulus (E") of the fiber-filled composites was higher than the control compound.
An increase of loss modulus with the addition of filler implies the superior viscous dissipatio n
property of the compounds than SBRC. Increased tear strength values in Table 2 also supported
this observation. The incorporation of the rigid solid raw fiber made the rubber stiff, and only a
slight positive shift of tan δ was observed. The addition of cellulose fiber also improved the wet
skid resistance of the fiber, whereas rolling resistance remained unaffected (Table S.3) (in
supporting information).
29
FIGURE 9. Effect of drumstick-derived micro and nanofibers on dynamic mechanical properties
of the composites.
SEM images of the cryofractured surface of the composites are presented in Figure 10 (a-f).
The images showed the microscale dispersion of the fibers in the rubber matrix. Micro-voids and
loosely bounded fibers were also visible in Figure 10.b. Alkali treatment increased surface
roughness and improved the rubber-microfiller interfacial interaction. Many uniformly distributed
small voids and pull-outs are visible throughout the matrix (Figure 10.c). Hence, AF microfiber
30
was homogeneously distributed in the SBR matrix. In the case of SBR5SF (as shown in Figure
10.d), the stearic acid-modified cellulose microfibers were uniformly dispersed in the rubber
matrix with less no of pull-outs and voids. The decrease in surface polarity of the SF improved the
rubber-microfiber interaction and enhanced the dispersion of the fiber in the SBR. Figure 10.f
gives a pictorial depiction of the dispersion of cellulose nanofiber in SBR. The size of the dispersed
nanofiber was in the range of a few micrometers. The hydrophilic nature and large surface area are
FIGURE 10. SEM images of the (a) SBRC, (b) SBR5RF, (c) SBR5AF, (d) SBR5SF, (e)
31
4. CONCLUSIONS
The successful separation of the cellulose micro and nanofibers from the drumstick fruit peel
and its application for the reinforcement of SBR is presented here. The percentage crystallinity
and the calculated diameter of the synthesized cellulose nanofiber were 62.3 % and 22 ±7 nm,
respectively. The enhanced surface roughness and decreased fiber diameter of the alkali-isolated
microfiber improved the mechanical properties of the composite compared to the raw drumstick
fiber-containing compound. Stearic acid and TESPT modification reduced the polarity of the
microfiber and enhanced the rubber-fiber compatibility. For the modified microfiber-containing
compounds, maximum bound rubber content was obtained for the stearic acid-modified
showed a maximum tensile strength of 1.9 MPa with improved elongation at break from 232 % to
301%. The bound rubber content, crosslink density and die swell index of all the nanofiber- loaded
compounds were better than the control compound. Hence, drumstick-extracted modified
cellulose microfibers and nanofibers of different compositions and polarity can be used as cost-
ACKNOWLEDGMENT
We would like to thank IIT Kharagpur for providing the fellowship to the student and research
FUNDING
This research did not receive any specific grant from funding agencies in the public, commercia l,
or not-for-profit sectors.
CONFLICT OF INTEREST
32
ORCID
REFERENCES
2. M. J. John, B. Francis, K. T. Varughese, S. Thomas. Compos. Part A Appl. Sci. Manuf. 2008,
https://doi.org/10.1016/j.carbpol.2012.10.021
https://doi.org/10.3390/jcs6070183
https://doi.org/10.1002/pc.20140
https://doi.org/10.1002/app.1995.070550405
https://doi.org/10.1016/j.carbpol.2007.12.005
33
10. B. Poyraz, A. Tozluoğlu, Z. Candan, A. Demir, M. Yavuz, Ü. Büyuksarı, H. İ. Ünal, H.
11. T. Saito, A. Isogai. Colloids Surf. A: Physicochem. Eng. Asp. 2006, 289, 219.
https://doi.org/10.1016/j.colsurfa.2006.04.038
15. M. M. Kabir, H. Wang, K. T. Lau, F. Cardona. Compos. B. Eng. 2012, 43, 2883.
https://doi.org/10.1016/j.compositesb.2012.04.053
18. N. Ayrilmis, F. Ozdemir, O. B. Nazarenko, P. M. Visakh. J. Compos. Mater. 2019, 53, 669.
https://doi.org/10.1177/0021998318789732
20. K. J. Sajithkumar, P. M. Visakh, E. V. Ramasamy. Waste and Biomass Valori. 2016, 5, 1227.
https://doi.org/10.1007/s12649-016-9499-z
https://doi.org/10.1007/s10570-006-9061-4
34
22. A. Sonia, K. P. Dasan. Carbohydr. Polym. 2013, 92, 668.
https://doi.org/10.1016/j.carbpol.2012.09.015
23. X. F. Sun, F. Xu, R. C. Sun, P. Fowler, M. S. Baird. Carbohydr. Res. 2005, 340, 97.
https://doi.org/10.1016/j.carres.2004.10.022
https://doi.org/10.3390/ma9040297
https://doi.org/10.3390/nano11020520
27. L. Y. Mwaikambo, N. Tucker, A. J. Clark. Macromol. Mater. Eng. 2007, 292, 993.
https://doi.org/10.1002/mame.200700092
28. X. Cao, Y. Chen, P. R. Chang, M. Stumborg, M. A. Huneault. J. Appl. Polym. Sci. 2008, 109,
3804. https://doi.org/10.1002/app.28418
https://doi.org/10.1016/j.carres.2010.10.020
https://doi.org/10.1016/j.carbpol.2011.06.030
31. H. Yang, R. Yan, H. Chen, D. H. Lee, C. Zheng. Fuel 2007, 86, 1781.
https://doi.org/10.1016/j.fuel.2006.12.013
35
32. S. Tsujiyama, A. Miyamori. Thermochim. Acta 2000, 351, 177.
https://doi.org/10.1016/S0040-6031(00)00429-9
33. W. Zhang, X. Zhang, M. Liang, C. Lu. Compos. Sci. Technol. 2008, 68, 2479.
https://doi.org/10.1016/j.compscitech.2008.05.005
https://doi.org/10.3390/molecules26030694
36. R. P. Kumar, M. L. G. Amma, S. Thomas. J. Appl. Polym. Sci. 1995, 58, 597.
https://doi.org/10.1002/app.1995.070580315
https://doi.org/10.1002/pc.750090306
https://doi.org/10.1002/pc.20332
41. A. Sinclair, X. Zhou, S. Tangpong, D. S. Bajwa, M. Quadir, L. Jiang. ACS Omega 2019, 4,
13189. https://doi.org/10.1021/acsomega.9b01313
369. https://doi.org/10.3390/polym12020369
36
43. D. Y. S. Low, J. Supramaniam, A. Soottitantawat, T. Charinpanitkul, W. Tanthapanichakoo n,
44. H. Kazemi, F. Mighri, K. W. Park, S. Frikha, D. Rodrigue. Polym. Compos. 2022, 43, 2362.
https://doi.org/10.1002/pc.26546
45. W. Jiang, P. Shen, J. Yi, L. Li, C. Wu, J. Gu. J. Appl. Polym. Sci. 2020, 137, 49163.
https://doi.org/10.1002/app.49163
46. B. Parambath Kanoth, M. Claudino, M. Johansson, L. A. Berglund, Q. Zhou. ACS Appl. Mater.
https://doi.org/10.1016/j.indcrop.2022.114935
49. P. Yu, H. He, Y. Luo, D. Jia, A. Dufresne. Macromolecules 2017, 50, 7211.
https://doi.org/10.1021/acs.macromol.7b01663
37