An Investigation on Moringa Oleifera Fruit (Drumstick)-Derived Cellulose Micro and

Nanofiber Reinforced Styrene Butadiene Rubber Composites

Riya Koley a, Uttam Kumar Ghoraib , Santanu Chattopadhyay a* , Anil K. Bhowmickc*

aRubber Technology Centre, Indian Institute of Technology Kharagpur, Kharagpur-721302, West Bengal, India
b Department of Industrial Chemistry and Applied Chemistry, Swami Vivekananda Research Center, Ramkrishn a
Mission Vidyamandira, Howrah-711202, West Bengal, India
cDepartment of Chemical and Biomolecular Engineering, The University of Houston, Houston, TX 77204-4004, USA

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

microfibers and nanofibers as sustainable fillers for high-performance elastomers composites.

KEYWORDS: Microfiber, cellulose nanofiber, Drumstick, rubber, Green-filler, composite.

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

with sustainable biobased or natural fillers.

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

Various cellulose micro/nanofiber separation processes like chemical treatment and

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

reported in the literature.10,11

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

works are available on cellulose nanofiber reinforcement of SBR-based composites. Sajithkumar

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

properties and processability of the SBR-based compounds is an interesting topic to study.

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

silane treatment, respectively. A comparative property study of the surface-modified microfiber-

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. MATERIALS AND METHODS

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

procured from Godrej Soap Limited, India. N-phenyl-N'-(1,3 dimethyl-butyl)-phenylenedia mine

(6PPD) and cyclohexyl benzothiazole sulfenamide (CBS) were supplied by National Organic

Chemicals Industries Ltd., India.

4
2.2. Methods

2.2.1. Isolation of raw cellulose fiber from drumstick peel (RF)

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%

with respect to the used raw cellulose fiber.

2.2.3. Stearic acid treatment of the drumstick-derived microfiber (SF)

The alkali-isolated microfibers were dipped in a 1 wt % stearic acid solution in a 60:40

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

microfiber for the modification.

2.2.4. TESPT modification of the drumstick-derived microfiber (SiF)

TESPT solution of 0.35 % concentration was prepared in a 90:10 ethanol-water mixture of

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

were dried in a lyophilizer and the yield was 32.5 %.

2.2.6. Preparation of SBR-drumstick fiber composite

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 .

Sixteen scans were set per sample.

6
 X-ray photoelectron spectroscopy (XPS) studies of the fiber were conducted in an Omicron

ESCA instrument, Oxford Instrument (Germany).

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

(S.Ⅱ) (in supporting information), respectively.

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

kV accelerating voltage. An average of 75-80 measurements is reported as the dimension of the

fibers.

Thermal transitions of the fibers were studied using a NETZSCH DSC 200 F3 Maia differentia l

calorimeter (DSC) in a nitrogen atmosphere. Approximately 10 mg of the powdered sample was

heated from 30 to 450 ˚C at a heating rate of 10 ˚C /min.

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

110 ˚C and 200-1500 s-1 shear rate.

The Cure characteristics of the rubber-fiber composites were measured with a rubber process

analyzer (D-RPA) at 150 ˚C temperature and 0.5˚ arc.

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

tensile specimens were punched out from the sheet.

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

performing one sided t-test.

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

calculated by using Equation (S.Ⅳ) (in supporting information).

Thermogravimetric analysis (TGA) of the raw drumstick fiber-containing compound, cellulose

microfiber and nanofiber-containing compounds was done in a Diamond TG/DTA PerkinElmer

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

3.1. FTIR analysis

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

cm-1 was for the C-O-C vibration of the pyranose structure23 .

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

nanofiber from the remaining lignin binder part.

FIGURE 1. FTIR spectroscopy of the drumstick-derived fibers.

3.2. XPS analysis

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.

A tiny peak for Si-O was also observed at 534.3 eV.

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

cellulose nanofiber (6.7%).

FIGURE 2. XPS of raw drumstick fruit fiber.

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

AF 0.70 1.8 19.0 35.4 45.6 - - 100

SF 0.62 2.2 24.5 57.2 4.3 8.4 5.6 86.7 13.3

SiF 0.69 5.9 24.9 53.5 21.6 - - 84.8 15.2

NF 0.66 0.7 6.7 6.3 87.0 - - 15.3 84.7

3.3. Morphology study of the isolated fibers

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

nanofiber of 30±10 nm diameter.28

14
FIGURE 3. Diameter and surface appearance of the RF (a. & b.), AF (c. & d.), SF (e. & f.), and

SiF (g. & h.).

15
 FIGURE 4. SEM (a & b) and TEM (c & d) images of the synthesized cellulose nanofiber (NF)

3.4. X-Ray diffraction analysis

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

the reason for the higher crystallinity percentage of the AF.

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

3.5. Comparative DSC study of the drumstick fibers

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

to the decrease in moisture evaporation temperature of the fibers.

Sulphuric acid hydrolyzed cellulose nanofibers have a sulfate functionalized surface, which

exceptionally lessens the moisture affinity of the cellulose. Therefore, moisture evaporation

endotherm for the drumstick nanofiber appeared at 86 ˚C.30

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

decrease in thermal stability of the isolated nanofibers.

3.6. Capillary rheology of the compounds

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

induced by fiber incorporation.

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

marginally lower than the control compound.

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

swell index of the compounds.

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

cellulose nanofiber-containing compounds were higher than the SBRC.

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

based on a one sided t-test.

3.7.1. Physico-mechanical properties of the cellulose micro/nanofiber-containing

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

of eucalyptus pulp-derived cellulose nanofiber, around 50 % improvement of the tensile strength

was reported for the SBR-based composite.38 Hence, the obtained improvement of tensile strength

of the drumstick nanofiber-containing compound was comparable to the previously reported

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

nanocrystal in a NR composite.46 But the properties of the fiber-reinforced composites cannot be

compared if the matrix is different.

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

ductility was obtained for SBR5NF.

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

3.8. Swelling study of the fiber-SBR composites

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

slightly higher than SBRC.

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.

3.9. Bound rubber content (BRC) of the composites

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

BRC value of the compounds increased.

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

increased. At higher loading, nanofibers started agglomerating, and rubber-filler interactio n

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.

3.10. TGA analysis

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

than the control compound.

FIGURE 8. Comparison of TGA curves of the fiber-loaded composites.

3.11. Dynamic mechanical analysis of the compounds

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.

3.12. SEM analysis

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

responsible for the agglomeration of nanofibers.

FIGURE 10. SEM images of the (a) SBRC, (b) SBR5RF, (c) SBR5AF, (d) SBR5SF, (e)

SBR5SiF and (f) SBR5NF.

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

microfiber-containing composite. Among the cellulose nanofiber-filled composites, the SBR5NF

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-

effective, sustainable fillers for the reinforcement of SBR composites.

ACKNOWLEDGMENT

We would like to thank IIT Kharagpur for providing the fellowship to the student and research

facilities for the work.

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

The authors have declared no conflicts of interest.

32
ORCID

Anil K. Bhowmick: http://orcid.org/0000-0002-8229-5353

REFERENCES

1. S. R. Djafari Petroudy, “Physical and Mechanical Properties of Natural Fibers” In Advanced

High Strength Natural Fibre Composites in Construction, Woodhead Publishing, Elsevier

Inc., 2017; pp 59–83. https://doi.org/10.1016/B978-0-08-100411-1.00003-0

2. M. J. John, B. Francis, K. T. Varughese, S. Thomas. Compos. Part A Appl. Sci. Manuf. 2008,

39, 352. https://doi.org/10.1016/j.compositesa.2007.10.002

3. N. Kasyapi, V. Chaudhary, A. K. Bhowmick. Carbohydr. Polym. 2013, 92, 1116.

https://doi.org/10.1016/j.carbpol.2012.10.021

4. H. Kazemi, F. Mighri, D. Rodrigue. J. Compos. Sci. 2022, 6,183.

https://doi.org/10.3390/jcs6070183

5. S. Shibata, Y. Cao, I. Fukumoto. Polym. Compos. 2005, 26, 689

https://doi.org/10.1002/pc.20140

6. V. G. Geethamma, R. Joseph, S. Thomas. J. Appl. Polym. Sci. 1995, 55, 583.

https://doi.org/10.1002/app.1995.070550405

7. D. Bhattacharya, L. T. Germinario, W. T. Winter. Carbohydr. Polym. 2008, 73, 371.

https://doi.org/10.1016/j.carbpol.2007.12.005

8. A. A. B. Omran, A. A. B. A. Mohammed, S. M. Sapuan, R. A. Ilyas, M. R. M. Asyraf, S. S.

R. Koloor, M. Petrů. Polymers 2020, 13, 231. https://doi.org/10.3390/polym13020231

9. A. Kumagai, N. Tajima, S. Iwamoto, T. Morimoto, A. Nagatani, T. Okazaki, T. Endo. Int. J.

Biol. Macromol. 2019, 121, 989. https://doi.org/10.1016/j.ijbiomac.2018.10.090

33
10. B. Poyraz, A. Tozluoğlu, Z. Candan, A. Demir, M. Yavuz, Ü. Büyuksarı, H. İ. Ünal, H.

Fidan, R. C. Saka. Fibers Polym. 2018, 19, 195. https://doi.org/10.1007/s12221-018-7673-y

11. T. Saito, A. Isogai. Colloids Surf. A: Physicochem. Eng. Asp. 2006, 289, 219.

https://doi.org/10.1016/j.colsurfa.2006.04.038

12. R. Koley, R. Kasilingam, S. Sahoo, S. Chattopadhyay, A. K. Bhowmick. Ind. Eng. Chem.

Res. 2019, 58, 18519. https://doi.org/10.1021/acs.iecr.9b03684

13. R. Koley, R. Kasilingam, S. Sahoo, S. Chattopadhyay, A. K. Bhowmick. Rubber Chem.

Technol. 2021, 94, 248. https://doi.org/10.5254/rct.21.79998

14. K. N. Bharath, P. Madhu, T. G. Y. Gowda, M. R. Sanjay, V. Kushvaha, S. Siengchin. J.

Polym. Environ. 2020, 28, 2823. https://doi.org/10.1007/s10924-020-01818-4

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

16. M. A. Kumar, G. R. Reddy, H. R. Rao, K. H. Reddy, B. H. N. Reddy. Int. J. Polym. Mater.

Polym. Biomater. 2012, 61, 759. https://doi.org/10.1080/00914037.2011.610046

17. K. Mishra, S. Sinha, Polym. Compos. 2020, 41, 5016. https://doi.org/10.1002/pc.25771

18. N. Ayrilmis, F. Ozdemir, O. B. Nazarenko, P. M. Visakh. J. Compos. Mater. 2019, 53, 669.

https://doi.org/10.1177/0021998318789732

19. S. Nayak, Polym. Compos.2022, 43, 6244. https://doi.org/10.1002/pc.26929

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

21. D. Bondeson, A. Mathew, K. Oksman. Cellulose 2006, 13, 171.

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

24. B. Ly, W. Thielemans, A. Dufresne, D. Chaussy, M. N. Belgacem. Compos. Sci. Technol.

2008, 68, 3193. https://doi.org/10.1016/j.compscitech.2008.07.018

25. K. Rieger, M. Porter, J. Schiffman. Materials 2016, 9, 297.

https://doi.org/10.3390/ma9040297

26. M. A. Gallardo-Sánchez, T. Diaz-Vidal, A. B. Navarro-Hermosillo, E. B. Figueroa-Ochoa,

R. Ramirez Casillas, J. Anzaldo Hernández, L. C. Rosales-Rivera, J. F. A. Soltero Martínez,

S. García Enríquez, E. R. Macías-Balleza. Nanomaterials 2021, 11, 520.

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

29. A. Kaushik, M. Singh. Carbohydr. Res. 2011, 346, 76.

https://doi.org/10.1016/j.carres.2010.10.020

30. A. Mandal, D. Chakrabarty. Carbohydr. Polym. 2011, 86, 1291.

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

34. U. Moonart, S. Utara. Cellulose 2019, 26, 7271. https://doi.org/10.1007/s10570-019-02611-

35. S. G. Chowdhury, J. Chanda, S. Ghosh, A. Pal, P. Ghosh, S. K. Bhattacharyya, R.

Mukhopadhyay, S. S. Banerjee, A. Das. Molecules 2021, 26, 694.

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

37. B. Karaağaç. Polym. Compos. 2014, 35, 245. https://doi.org/10.1002/pc.22656

38. M. A. Raza, M. A. Ashraf, A. V. K. Westwood, T. Jamil, R. Ahmad, A. Inam, K. M. Deen.

Polym. Compos. 2016, 37, 1113. https://doi.org/10.1002/pc.23273

39. L. I. Rueda, C. C. Antón, M. C. T. Rodriguez. Polym. Compos. 1988, 9, 198.

https://doi.org/10.1002/pc.750090306

40. M. Haghighat, S. N. Khorasani, A. Zadhoush. Polym. Compos. 2007, 28, 748.

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

42. K. Roy, S. C. Debnath, L. Tzounis, A. Pongwisuthiruchte, P. Potiyaraj. Polymers 2020, 12,

369. https://doi.org/10.3390/polym12020369

36
43. D. Y. S. Low, J. Supramaniam, A. Soottitantawat, T. Charinpanitkul, W. Tanthapanichakoo n,

K. W. Tan, S. Y. Tang. Polymers 2021, 13, 550. https://doi.org/10.3390/polym13040550

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.

Interfaces 2015, 7, 16303. https://doi.org/10.1021/acsami.5b03115

47. V. Balachandrakurup, J. Gopalakrishnan. Ind. Crops Prod. 2022, 183, 114935.

https://doi.org/10.1016/j.indcrop.2022.114935

48. M. Dominic, R. Joseph, P. M. Sabura Begum, B. P. Kanoth, J. Chandra, S. Thomas.

Carbohydr. Polym. 2020, 230, 115620. https://doi.org/10.1016/j.carbpol.2019.115620

49. P. Yu, H. He, Y. Luo, D. Jia, A. Dufresne. Macromolecules 2017, 50, 7211.

https://doi.org/10.1021/acs.macromol.7b01663

50. E. Abraham, B. Deepa, L. A. Pothan, M. John, S. S. Narine, S., Thomas, R. Anandjiwala.

Cellulose 2013, 20, 417. https://doi.org/10.1007/s10570-012-9830-1

37