Graphic abstract

figure a

Introduction

Cellulosic polymer nanocomposites are systems in which cellulose-derived nanofillers are incorporated into a polymer with the mission to enhance thermal and mechanical properties (Khattab et al. 2017). Cellulosic nanofibrillar structures have been at the core of attention over the past two decades in view of being plentiful and renewable. The main sources from which cellulose nanofibrils are obtained include in wood, agricultural residues, and bacterial cellulose. Cellulose nanocrystals (CNCs) are known for their biodegradability, relatively high strength, and Young’s modulus; therefore, CNCs have been widely utilized in the development of cellulosic nanocomposites (Mariano et al. 2014). Compared to classical fibers, CNCs have higher aspect ratio, surface area, higher crystallinity, flame retardancy, and thermal stability (Foster et al. 2018). They have elastic modulus (139.5 GPa) and Young’s modulus (206 GPa) values more or less equal to those of Kevlar® and steel (Lin and Dufresne 2014). Nevertheless, poor properties of cellulosic nanocomposites are addressed as a consequence of the lack of compatibility between cellulose and plastics (Hubbe et al. 2008). In this sense, CNC modification/functionalization and CNC separation from new sources have attracted widespread interests in academia and industry alike.

There are several reports on the preparation and characterization of elastomer-based nanocomposites reinforced with nanocellulose. A variety of rubbers including natural rubber (NR), styrene-butadiene rubber (SBR), epoxidized natural rubber (ENR), and butadiene rubber (BR) have been reinforced for the thermal, mechanical, barrier, and dynamic mechanical properties (Nunes 2017). Flauzino Neto et al. studied the reinforcing effect of soy hull-derived CNCs in NR and reported that the improvement in the mechanical properties of the composites was greatly influenced by the aspect ratio and percolating network of CNCs (Flauzino Neto et al. 2016). Chen et al. investigated the role of CNCs in the cure behavior, cross-link density, and dynamic mechanical properties of NR/BR/SBR blends (Chen et al. 2014). They observed that CNCs can be used as a green alternative for carbon black (at 10 phr loading) in NR/BR/SBR blend when resorcinol and hexamethylenetetramine (RH) was used to modify the interface. Visakh et al. evaluated the transport properties of cellulosic nanofibers (CNFs) and crystals (CNCs) reinforced natural rubber composites. They observed that the equilibrium solvent uptake and swelling index were found to be minimum for CNF reinforced NR composites rather than CNCs (Visakh et al. 2011).

Acrylonitrile butadiene rubber (NBR) is a polar rubber that enjoys from high compression set, heat resistance, oil resistance, and abrasion resistance (El-sabbagh and Mohamed 2011). In this regard, it has been considered in manufacturing diverse products such as seals, hoses, pipelines, belts, grommets, mats, and gloves ( Taib et al. 2017). However, it is not biodegradable and duffers from lack of strain-induced crystallization (Sadeghalvaad et al. 2019a). NBR has been reinforced with CNCs in several studies. Taib et al. improved the tensile strength of NBR from 8.26 to 16.58 MPa (ca. two-fold) by incorporation of 5 phr CNCs (Taib et al. 2020). Cao et al. observed the ‘Payne effect’ in CNC-reinforced NBR composites (Cao et al. 2013a). They reported that the glass transition temperature (Tg) improved from 10.8 to 17.2 °C with the incorporation of 20 phr CNCs into NBR. Chen et al. reported enhanced modulus at −70 °C of NBR foam from 610 to 1150 MPa (ca. two-fold) by incorporating 15 phr CNCs (Chen et al. 2014). They concluded that the hydrogen bonding between the hydroxyl groups of CNCs and nitrile groups of NBR is responsible for higher performance. There have been also some reports on surface modification of CNCs, e.g. with gallic acid to prepare antioxidant CNC (Aox-CNC) which improved tensile properties, thermal stability, and biodegradability of NBR (Taib et al. 2019). However, in most cases highly loaded cellulosic NBR nanocomposites were prepared, which arises from inadequate properties of CNCs. This highlights the need for finding proper sources of CNCs with high interfacial adhesion to NBR at low loading levels.

The incompatibility between the polar filler and the non-polar matrix is the main reason for poor properties. Surface modification of matrix or filler with suitable surface modifying/coupling agents is a key to improve interfacial adhesion. Acetylation, urethanization, etherification, peptide coupling, esterification, silylation, titanate coupling, polymer grafting, and 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO) mediated oxidation are some of the common ways to improve the hydrophobicity of CNFs and thereby enhance the interfacial adhesion between CNFs and the non-polar polymer matrix (Chakrabarty and Teramoto 2018). It is reported that the use of resorcinol-hexamethylenetetramine (RH) could improve the dispersion of CNC in natural rubber (NR) (Jiang and Gu 2020). It is noticed that the rolling resistance of NR/CNC/RH composite was less than that of NR/CNC composite without compromising the wet-skid resistance, which is crucial for developing green tires. CNFs derived from Agave angustifolia can improve the thermal stability and biodegradability of polylactic acid/natural rubber blend when compatibilized with liquid natural rubber (Rosli et al. 2019). Eldho et al. studied the influence of the Zn-cellulose complex and three-dimensional network structure of CNFs in improving the mechanical and dynamic-mechanical properties of natural rubber (Abraham et al. 2013a). Attachment of –SH or –C = C functional groups to the surface of CNFs by grafting it with suitable surface modifying agents enhanced solvent (toluene) uptake, and rubber damping is greatly decreased by the introduction of grafted CNFs in styrene-butadiene rubber (SBR) (Sinclair et al. 2019). The influence of hydrogen bonding between the filler and the matrix in improving the mechanical properties of the composites was well explained in silica-NBR (Suzuki et al. 2005), cellulose-NBR (Chen et al. 2015), and halloysite nanotube-NBR systems (Ismail and Ahmad 2013).

The use of coupling agents is one of the classical ways to improve the interfacial adhesion between the filler and the polymer. But, the utilization of costly coupling agents (e.g. silane coupling agents) in improving the dispersibility of CNFs in the polymer matrix makes the entire process tedious and uneconomic. Herein, we incorporated the cellulose nanofibers in the NBR latex to prepare a masterbatch so as to enhance the dispersibility and interfacial adhesion between CNFs and the NBR. Then, the masterbatch was compounded with solid NBR and other vulcanizing agents in a two-roll mill to improve the properties of NBR. We make use of hydrogen bonding interaction between the –OH groups of CNFs and –CN groups of NBR as a simple yet smart way to improve the physico-chemical, thermal, mechanical, transport, and dynamic mechanical properties. To the best of the knowledge of the authors of this work, no detailed investigation is available on the impact of the latex coagulation process followed by two rolls milling in processing NBR/CNF composites. The new source of CNFs (Cuscuta reflexa) and the lower amount of CNFs (2 to 6 phr) used in this study compared to previous works are features of innovative approach developed. Moreover, there is no detailed investigation on the reinforcing effect of plant-derived CNFs on the NBR. The Cuscuta reflexa is a parasitic plant consisting of 100–170 species of yellow, orange, or red parasitic plants mainly found throughout the temperate and tropical regions of the world, and commonly available in Kerala, India in large volume (Paul et al. 2019). In a previous work, we used CNFs with high crystallinity index (67 %) and diameter in the range 10–30 nm extracted from the parasitic plant Cuscuta reflexa using steam explosion in presence of 5 % oxalic acid and found it a promising green nanofiller with acceptable compatibility with elastomers (Dominic C.D. et al. 2020a). Herein, NBR/CNF nanocomposites were prepared and characterized for cure behavior, physico-mechanical, thermal, fractographic, viscoelastic, and transport properties. Attempts have been paid to emphasize the significance of the careful selection of cellulosic fibers and elastomer matrix which guarantees adequate interfacial interaction for high-performance applications.

Materials and methods

Materials

The cellulosic nanofibers (10–30 nm) with high crystallinity index (67 %) isolated from Cuscuta reflexa were used in this study (Dominic et al. 2020a). The plant possesses a distinct appearance being leafless, rootless, and smooth lacking chlorophyll can parasitize the host through haustoria. Agriculturalists consider them as a destructive weed as it can parasitize multiple adjacent hosts simultaneously which can also affect commercially valuable crops such as flax, alfalfa, bean, and potatoes. The chopped Cuscuta reflexa fibers were heated with 2 % NaOH solution to remove hemicellulose and lignin. The alkali-treated fibers were steam-exploded in an autoclave with 5 % oxalic acid to yield CNFs. The detailed procedure for the extraction of CNFs and their characterization are described in our previously reported work (Dominic et al. 2020a). Acrylonitrile-butadiene latex with 34 % acrylonitrile (AN) content was obtained from Eliokem India Pvt. Ltd., Bombay. Acrylonitrile-butadiene rubber (KNB 35 L) with 34 % AN content was purchased from Kumho Petrochemicals Co., Ltd., South Korea. The rubber additives used such as zinc oxide, stearic acid, 1,2-dihydro-2,2,4-trimethyl quinoline (TQ), N-cyclohexyl-2-benzothiazole sulphenamide (CBS), tetramethylthiuram disulphide (TMTD), sulphur, oxalic acid, hydrochloric acid, sodium hydroxide, hydrogen peroxide, and calcium chloride were industrial-grade used without further purification.

Methods

Preparation of NBR/CNF composites

The rubber compounding was done according to ASTM D 3184. The NBR/CNF composites were prepared as follows. First, the masterbatches of NBR latex/CNF were prepared by coagulation using a 0.5 % CaCl2 solution. This was followed by compounding the masterbatches of NBR/CNF latex with solid NBR using a two-roll mill (size 6″ × 12″). The compound was then mixed with vulcanizing agents and other ingredients. Later, the vulcanization was carried out using an electrically heated hydraulic press at 160 °C and a pressure of 2844 psi. The formulations of the prepared NBR/CNF composites are given in Table 1.

Table 1 Formulations for NBR/CNF composites. In this table, formulations are based on 100 parts by weight of NBR and other ingredients are added in part per hundred (phr), i.e. part by weight of additives based on 100 parts by weight of reference NBR
Full size table

Characterization

The rubber process analyzer, RPA 2000 was used to study the cure characteristics of NBR/CNF composites based on ASTM D 5289. The tensile properties and tear resistance of the composites were measured using a universal testing machine (Shimadzu Model AGI) based on ASTMD 412 and ASTM D 624, respectively. Shore A type durometer was used to study the hardness of NBR and NBR/CNF composites based on ASTM D 2240. Bareiss DIN abrader, Germany was used to study the abrasion resistance according to ASTM D 5963. The compression set apparatus was used to study the compression set of the NBR composites according to ASTM D 395. TA instrumentQ-50 thermogravimetric analyzer was used to measure the thermal properties of the composites. The test was carried out in a nitrogen atmosphere at a heating rate of 20 °C /min from 30 to 500 °C. Scanning electron microscope JEOL JSM 8390 LV was used to study the tensile fracture surface of the NBR/CNF composites.

Swelling behavior and cross-link density of the NBR/CNF composites were studied as in our previous publication (Dominic et al. 2020b). The equilibrium swelling method was used to evaluate the swelling behavior of the composites. Toluene was used as the solvent.

The solvent uptake (Qt), swelling coefficient, and swelling index were measured using the Eqs. 1, 2 and 3:

$${Q}_{t}=\left(\frac{{(\text{W}}_{\text{S}}-{\text{W}}_{\text{i}})/{\text{M}}_{\text{s}}}{{\text{W}}_{\text{i}}}\right)\times 100\text{\%}$$
(1)
$$Swelling\, Coefficient=\frac{{W}_{s}-{W}_{i}}{{W}_{i}}\times {\rho }_{s}$$
(2)
$$Swelling\, Index=\frac{{W}_{s}-{W}_{i}}{{W}_{i}}$$
(3)

where Wi and Ws are the weight of the composite samples before and after the swelling studies. Ms and \({\rho }_{s}\) are the molar mass and density of the solvent used (toluenē).

The cross-link density (ν) of the nanocomposites was calculated using Eq. (4):

$$\nu =\frac{1}{2{M}_{c}}$$
(4)

where Mc is the molecular weight of NBR segments between the cross-links. Based on the Flory-Rehner theory, the Mc is given by Eq. 5 (Flory and Rehner 1943):

$${M}_{c}=\frac{-{\rho }_{r}{V}_{s}{V}_{r}^{1/3}}{\text{ln}\left(1-{V}_{r}\right)+{V}_{r}+\chi {V}_{r}^{2}}$$
(5)

In this equation ρr is the density of the NBR, Vs is the molar volume of toluene, Vr is the volume fraction of swollen NBR and χ is the NBR-toluene interaction parameter (Bryan and G. N. Welding 1964).

The values of Vr and χ were calculated using the Eqs. 6, 7.

$${V}_{r}=\frac{(d-fw){\rho }_{r}^{-1}}{\left(d-fw\right){\rho }_{r}^{-1}+{A}_{s}{\rho }_{s}^{-1}}$$
(6)
$$\chi =\beta +\frac{{V}_{s}{({\delta }_{s}-{\delta }_{p})}^{2}}{RT}$$
(7)

In this equation, d is the deswollen weight of the NBR, f is the volume fraction of the insoluble components, w is the initial weight of the NBR, \(\beta\) is the lattice constant, R is the gas constant, T is the absolute temperature, and δs and δp are the solubility parameter of the toluene and NBR.

TA Instrument DMA Q800 was used to study the viscoelastic properties of the NBR/CNF composites in tension mode from −40 to 80 °C. The sample dimension used for the study was 30 × 3 × 2 mm3. The composites were tested at a heating rate of 3 °C/min. The frequency used for testing was 1 Hz.

The volume fraction of the NBR chains immobilized because of the filler-polymer interactions was determined from the DMA studies using Eq. 8 (Formela et al. 2016):

$${C}_{v}=1-\left(1-{\text{C}}_{0}\right)\frac{\text{W}}{{\text{W}}_{0}}$$
(8)

where \({C}_{v}\)and C0 are the volume fraction of the constrained regions in NBR/CNF composites and pure NBR, respectively. The value of C0 is taken as zero. The\(\text{W}\) and \({\text{W}}_{0 }\)are the energy loss fraction of CNF reinforced NBR composite and neat NBR. The energy loss fraction was related to tanδ by the following Eq. (9):

$$W=\frac{{\uppi }\text{tan}{\updelta }}{{\uppi }\text{tan}{\updelta }+1}$$
(9)

Further, to study the interfacial adhesion between the nanofiber and NBR matrix, the adhesion factor () was evaluated using Eq. 10 (Formela et al. 2016):

$$W=\frac{{\uppi }\text{tan}{\updelta }}{{\uppi }\text{tan}{\updelta }+1} \frac{\text{tan}{{\updelta }}_{\text{c}}}{\text{tan}{{\updelta }}_{\text{m}}} -{\text{V}}_{\text{f}} +A$$
(10)

Where tan δc and tan δm are the loss tangent of CNF reinforced NBR composites and pure NBR, respectively and \({V}_{f}\) is the volume fraction of the CNFs.

Results and discussion

Fourier Transform Infrared (FTIR) analysis

The FTIR analysis was conducted to study the interaction between CNFs and NBR. The FTIR spectra of pure CNFs, pristine NBR and NBR/CNF4 composite are shown in Fig. 1. The functional group analysis of pure CNFs is detailed in our previously reported work (Dominic et al. 2020). The transmission peaks of NBR observed at 2921 − 2846 cm− 1 which correspond to asymmetric and symmetric –CH vibration (Sadeghalvaad et al. 2019b). The peaks observed at 2237 cm− 1 and 1719 cm− 1 in the FTIR spectra of neat NBR correspond to –CN stretching and –C = O stretching vibrations (Abdul Rashid et al. 2018). The peak at 967 cm− 1 is related to C-H stretching of butadiene double bond (–CH = CH (trans)) and the peak at 690 cm− 1 corresponds to the RCH = CHR bending vibration (Abdul Rashid et al. 2018). The peaks located at 1440, 1350, 1064 cm− 1, and 1540 cm− 1 in the FTIR spectra of pristine NBR indicate –CH2 bending vibration, –CH2 deformation, C-O stretching, and -C = C- bond stretching respectively (Ramesan and Nihmath 2017, Abdul Rashid et al. 2018). The increase in the intensity of the peak at 1089 cm− 1 in the FTIR spectra of the NBR/CNF4 composite confirms the presence of CNFs in the NBR matrix. The peak at 1089 cm− 1 corresponds to the glycosidic rings of CNF chains (Abdul Rashid et al. 2018). The peak positioned at 2237 cm− 1 is transit to 2238 cm− 1 (higher wavenumber region) in the FTIR spectra of NBR/CNF4 composite indicate effective interaction has taken place between the –CN functional groups of NBR and –OH functional groups of CNFs. A similar observation was also made by Sadeghalvaad et al. while studying the interaction between nanoclay, CaCO3 and silica with NBR (Sadeghalvaad et al. 2019b). The shift in the position of the peak from 1540 cm− 1 to 1602 cm− 1 in the FTIR spectra of NBR/CNF4 composite also confirms the interfacial interaction between CNFs and NBR.

Fig. 1
figure 1

FTIR spectra of pure CNFs, pristine NBR, and NBR/CNF4 composite

Full size image

X-ray diffraction analysis (XRD)

The XRD analysis was performed on pristine NBR and NBR/CNF composites to evaluate the phase purity and fiber-matrix interaction. The XRD pattern of pure CNFs is shown in Fig. 2 A. The XRD analysis of pure CNFs is detailed in our previously reported work (Dominicet al. 2020). Figure 2B showed the XRD patterns of pristine NBR and NBR/CNF composites. The pristine NBR shows a broad peak at 2θ = 19.47° is due to the amorphous nature of the polymer matrix. All the prepared composites also show a broad diffraction peak at around 2θ = 19°. The broad diffraction peaks for NBR/CNF2, NBR/CNF4, and NBR/CNF6 composites are observed at 19.15°, 19.07°, and 18.99° respectively. The shift in diffraction peak to lower 2θ values for the composites indicate good interaction between the CNFs and the NBR matrix. On the other hand, the decrease in the breadth of the peaks in the composite samples indicate a reduction in the amorphous nature of the NBR polymer. Ramesan and Nihmath observed a similar shift in 2θ values and a decrease in the breadth of the main peak in NBR-hydroxyapatite nanocomposites (Ramesan and Nihmath 2017). Apart from the broad diffraction peak centered at around 2θ = 19°, a shoulder peak at 2θ = 22.27° is observed in all NBR/CNF composites. The peak at 2θ = 22.27° is the characteristic crystalline diffraction peak from the (200) plane of cellulose nanofibers (Bahloul et al. 2021). This peak is not observed in the XRD pattern of pristine NBR. The presence of the shoulder peak at 2θ = 22.27° in NBR/CNF composites confirms the successful incorporation of CNFs in to the NBR matrix.

Fig. 2
figure 2

a Diffraction patterns of pure CNFs b pristine NBR and NBR/CNF composites

Full size image

Cure characteristics of NBR/CNF composites

Table 2 Curing parameters and Wolff activity coefficient determined for NBR/CNF nanocomposites
Full size table

Table 2 exhibits the cure characteristics of CNF reinforced NBR composites. The optimum cure time is the time required to reach 90 % of the maximum torque and the values of optimum cure time of NBR/CNF composites are observed to be higher than NBR. The interaction of the hydroxyl groups of cellulosic nanofibers with accelerators and activators may reduce the number of active curing agents required for vulcanization. That might be the reason for the increase in cure time for the prepared NBR/CNF nanocomposites. Rajkumar et al. also reported an increase in cure time when silica particles are used as a reinforcing filler in NBR (Rajkumar et al. 2013). Scorch time is the time required for the onset of vulcanization, or in other words, it gives information on the processing safety of the prepared nanocomposites. It was observed that scorch time increases with an increase in CNF loading. Thus, the addition of CNFs into the NBR matrix improves the scorch safety of the prepared NBR-CNF nanocomposites. The cure rate index (CRI) of the prepared nanocomposites is found to be decreasing with an increase in CNF content. Among the nanocomposites prepared, NBR/CNF4 shows the highest CRI value. The low cure rate index makes the processability of the composite difficult. The decrease in the CRI of NBR/CNF composites was probably due to the restricted movement of free radicals generated during vulcanization by the surface functionalities (hydroxyl groups) of CNFs (Kulshrestha et al. 2020). Thus, the cellulosic nanofibers have a negative effect on the cure reaction of NBR. The minimum torque provides information on the viscosity of the material and is higher for the prepared NBR-CNF nanocomposites compared to pristine NBR. The maximum torque which is an indication of the shear modulus of the material was found to be maximum for NBR/CNF4 composite. The increase in the maximum torque values of NBR/CNF composites is an evidence of the confinement of the polymer chains and reduced macromolecular mobility. The values of maximum torque is dependent on the cross-link density and fiber-matrix interaction (Visakh et al. 2012b). The maximum torque for NBR/CNF4 composite shows better entanglement of rubber chains with CNFs. At high CNF loading, the agglomeration of nanofibers results in poor wetting of the fibers with polymer chains. This is the reason for the decrease in the maximum torque at higher filler loading.

The activity of CNFs in NBR can be monitored by the change in torque moment when the sample is cured in an oscillatory rheometer. The Wolff activity coefficient (\({\alpha }_{F})\) gives information on the reinforcing effect of filler in the matrix. Studies have shown that the composites with a higher valueof \({\alpha }_{F}\) shows better reinforcing effect (Formela et al. 2016). Wolff activity coefficient depends on the parameters \({\varDelta M}_{x}\) (the torque increment of NBR/CNF composites) and \({\varDelta M}_{0}\) (torque of NBR)\(,{m}_{x}\) (the weight of CNFs) and \({m}_{p}\) (the weight of NBR matrix). The value \({\alpha }_{F}\) gives information on the effect of reinforcement of the fibers in the NBR matrix. The negative value showed no reinforcement, while the positive value showed good reinforcement. The value of \({\alpha }_{F}\) depends on the total composition of the rubber compound, type of the rubber, nature of the filler and the curing system (Kruželák et al. 2015).The \({\alpha }_{F}\) can be calculated from the following Eq. (11):

$$\frac{{\varDelta M}_{x}}{{\varDelta M}_{0}}-1={\alpha }_{F}\frac{{m}_{x}}{{m}_{p}}$$
(11)

where \({\varDelta M}_{x}\) and \({\varDelta M}_{0}\) are the torque increment of NBR/CNF composites and NBR, respectively. The \({ m}_{x}\) and \({m}_{p}\)are the weight of CNFs and NBR matrix. The values of \({\alpha }_{F}\) is positive for all the NBR/CNF composites, this shows the reinforcing effect of CNFs with NBR matrix. From Table 2, the maximum value of \({\alpha }_{F}\) was observed at 2 phr CNF reinforced NBR composites due to the improved reinforcing effect of nanofiber with NBR matrix at this composition. The values of \({\alpha }_{F}\) obtained for NBR/CNF composites were found to be greater than \({ \alpha }_{F}\) values obtained for chitin nanowhiskers reinforced NBR composites (Dominic et al. 2020b).

Mechanical properties of NBR/CNF composites

Table 3 Physico-mechanical and sorption parameters of NBR/CNF nanocomposites
Full size table

The tensile properties of CNF reinforced NBR composites are given in Table 3. The prepared NBR/CNF composites show higher tensile strength compared to neat NBR. The NBR/CNF4 composite shows ~ 125 % improvement in tensile strength compared to neat NBR. The stress-strain curves of the prepared NBR/CNF composites are shown in Fig. 3 A. The stress increases linearly with strain initially for all the prepared composites and rubbery deformation happens in the later stage. The improvement in the tensile strength of the prepared nanocomposites can be explained based on the stress transfer theory. At low CNF loading (up to 4 phr), the well-dispersed nanofibers favor effective stress transfer between the matrix and CNFs, and thus helps to prevent the propagation of microcracks. The hydrogen bonding between CNFs and NBR acts as a network, which also helps to transfer the stress continuously from the matrix to the filler. The improvement in the tensile strength of the prepared nanocomposites is in line with the ‘Chinese finger trap effect’ (Wisittanawat et al. 2014). Thus, the hydrogen bond formation between the hydroxyl groups of nanofibers and the nitrile groups of NBR at the interface, good filler dispersion, and high aspect ratio of the nanofibers led to the improvement in tensile strength of the NBR-CNF composites (Abdul Rashid et al. 2018). The improvement in the tensile strength (~ 125 %) of NBR reinforced with 4 phr of CNFs was observed to be greater than the improvement (~ 116 %) in tensile strength reported by the incorporation of 2 phr chitin nanowhiskers (Dominic et al. 2020b). The deterioration in tensile strength after 4 phr CNF loading was attributed to the agglomeration of nanofibers. At high CNF loading, the nanofibers tend to aggregate and form bundles. Then the fiber-fiber interaction will be more than the fiber-matrix interaction which restricts the effective stress transfer from the matrix to the fiber. The reduced crosslink density and dispersion issue of CNFs are responsible for the low tensile strength of the NBR/CNF6 composite (Cao et al. 2013b)

The elongation at break of NBR, NBR/CNF2, NBR/CNF4, NBR/CNF6 composites was found to be 398 ± 24 %, 321 ± 14 %, 314 ± 28 %, 277 ± 21 % respectively. The nanocomposites show lower elongation at break compared to neat NBR. The modulus at 100 % elongation of NBR, NBR/CNF2, NBR/CNF4, and NBR/CNF6 composites are observed to be 1.07 ± 0.07 MPa, 1.80 ± 0.09 MPa, 1.87 ± 0.04 MPa, and 1.84 ± 0.08 MPa respectively. The increase in the modulus of the NBR/CNF nanocomposites is attributed to the improved surface activity and rubber-nanofiber interaction. The confinement of polymer chains by the well-dispersed cellulosic nanofibers provides more stiffness which in turn results in improved modulus values of the prepared NBR/CNF nanocomposites. The modulus drop at high CNF loading was probably due to the agglomeration of nanofibers and lower cross-link density (Cao et al. 2013a). The tear strength of NBR, NBR/CNF2, NBR/CNF4, and NBR/CNF6 are observed to be 14.54 ± 0.23 N/mm, 28.54 ± 0.56 N/mm, 29.84 ± 0.34 N/mm, and 27.08 ± 0.12 N/mm respectively. About 105 % improvement in tear strength was obtained by introducing 4 phr of CNFs in the NBR matrix. The increase in the tear strength is due to the change in the nature of the tear process from a smooth and straight path to a rougher tortuous path. The well-dispersed cellulosic nanofibers can obstruct the tear propagation and more energy is needed to separate the two surfaces which in turn results in higher tear strength. The lower value of the tear strength for sample with 6 phr of CNFs is probably due to the agglomeration of nanofibers. The tensile analysis shows that the dispersion of the cellulose nanofibers in the NBR latex plays a major role in improving the mechanical properties of the prepared composites. By latex stage processing, CNFs can be uniformly dispersed in the NBR matrix which in turn results in effective stress transfer. When the fibers are well dispersed and oriented in the mill direction, they can act as effective stress carriers owing to the superior mechanical properties of the prepared composites (NBR/CNF4). At higher fiber loading (6 phr), the fiber-fiber agglomeration results in the deterioration of the mechanical properties

The hardness (Shore A) of the prepared nanocomposites is improved considerably compared to neat NBR and the maximum hardness is reported for NBR/CNF4 composite. This suggests that the NBR-CNF4 composite offers more resistance to the reversible deformation by a rigid indenter. The improvement in the hardness of the NBR/CNF4 composite is probably due to the increased stiffness and cross-link density (Sae-Oui et al. 2002). At higher CNF loading, agglomeration of fibers results in lower crosslink density, which may be detrimental to hardness. The abrasion resistance index is highest for NBR/CNF4 composite. About 27 % improvement in the abrasion resistance index is obtained for NBR/CNF4 composite compared to neat NBR. The good dispersion of CNFs, high hardness, and high cross-link density of NBR/CNF4 composite prevents the removal of the matrix by the DIN abrader (Sombatsompop et al. 2004). Among the composites, the abrasion loss is highest for the NBR/CNF6 composite because of the agglomeration of the nanofibers and low cross-link density

It is observed that the compression set values of the prepared nanocomposites increase with an increase in CNF loading. A reverse trend was observed when organoclay, silica, and calcium carbonate were used as fillers in NBR (Sadeghalvaad et al. 2019a). Among the prepared nanocomposites, the compression set is observed to be a minimum for the NBR/CNF2 composite. This suggests that the deformations due to the applied pressure force to NBR/CNF2 composite are greatly reduced by the introduction of well-dispersed CNFs in the polymer matrix. The high compression set value of the NBR/CNF6 composite is attributed to the agglomeration of the nanofibers (Rattanasom et al. 2007)

Swelling studies

The transport of solvent molecules across the NBR matrix depends on the parameters such as the free volume present in the NBR or NBR nanocomposite, interfacial interaction between the nanofibers and NBR, cross-link density, the segmental mobility of the NBR chains, the size of the penetrant molecule, type of the filler/fiber and temperature. The swelling studies provide information on the interfacial adhesion and reinforcing effect of CNFs in NBR. The graphs of solvent uptake (Qt) with respect to time for NBR and NBR/CNF composites are shown in Fig. 3B. All the prepared samples show a rapid increase in solvent uptake (because of concentration gradient) initially and an equilibrium plateau is reached on the later stage. From Fig. 3B, the sorption of the solvent is lowest at NBR/CNF4 composite. The well-dispersed CNFs can form physical and chemical cross-links with NBR polymer chains which result in decreased chain flexibility that in turn restrict the entry of solvent to the NBR matrix (Moni et al. 2018).

The sorption parameters such as swelling index and swelling coefficient of NBR/CNF composites are given in Table 3. The swelling index of NBR, NBR/CNF2, NBR/CNF4, and NBR/CNF6 composites are observed to be 204.24 ± 0.5, 196.59 ± 0.7, 196.43 ± 0.50, and 206.21 ± 0.60 respectively. The swelling index is minimum for sample reinforced with 4 phr of CNFs. This is due to the decrease in the free volume for this system related to the strong interfacial interaction between CNFs and NBR matrix. At 4 phr CNF loading, the well-dispersed nanofibers can bring the chains closer and keep them intact which in turn results in a low swelling index value of the NBR/CNF4 composite. A similar decrease in swelling index was observed when chitin nanowhiskers were used as a filler in NBR (Dominic et al. 2020). The swelling coefficient of NBR, NBR/CNF2, NBR/CNF4, and NBR/CNF6 composites are found to be 1.78 ± 0.02, 1.71 ± 0.01, 1.70 ±0.01, and 1.79 ± 0.02 respectively. The tortuous path, tangling effect and the three-dimensional networks created by the cellulosic nanofibers restrict the entry of non-polar solvent molecules into the NBR matrix. Thus, the segmental mobility of the polymer chains and the movement of the solvent molecule between the fiber and polymer interface and polymer-polymer interface get restricted owing to the low value of the swelling coefficient in the prepared NBR/CNF composites except for NBR/CNF6. At high CNF loading (6 phr), CNFs get agglomerated and form bundles. Then the fiber-fiber interaction will be more than the fiber-matrix interaction owing to the high swelling coefficient of NBR/CNF6 composite. The hydrogen bonding between the nitrile rubber and nanofibers may also increase the effective cross-link density, which is also a crucial parameter that influences swelling. The cross-link density is defined as the number of cross-links per unit volume. Higher the cross-link density, lower will be the degree of swelling (Hoti et al. 2021; Vural et al. 2010). The cross-link density (× 10−4 mol/cc) of NBR, NBR/CNF2, NBR/CNF4, and NBR/CNF6 composites are found to be 1.51 ± 0.11, 1.63 ± 0.13, 1.65 ± 0.12, and 1.34 ± 0.08 respectively. The obtained results of cross-link density of NBR/CNF composites are in line with the swelling index measurements. The cross-link density is maximum for NBR/CNF4 composite. The complex network structure formed by the fiber-fiber and fiber-polymer interactions increased the cross-link density of the material which in turn restricts the entry of solvent molecules to the polymer matrix (Abraham et al. 2013b). It is observed that the cross-link density of the NBR/CNF6 composite is less than that of neat NBR. The decrease in the cross-link density at high CNF loading is attributed to the agglomeration and dispersion issue of cellulosic nanofibers. Boonbumrung et al. reported a similar observation in the cross-link density when MWCNTs are introduced as filler in NBR (Boonbumrung et al. 2016). The swelling analysis shows that the sorption parameters could be considered to predict the physico-mechanical performance of the prepared composites. The reduced segmental mobility of polymer chains is a direct consequence of the effective interaction between the filler and the matrix which in turn results in superior mechanical performance. When the fibers are uniformly distributed in the polymer matrix, the free volume space decreases. The decrease in the free volume space together with the reduced segmental mobility of the polymer chains restrict the entry of solvent to the composite owing to a lower swelling coefficient. Here, the lowest swelling coefficient is reported for NBR/CNF4 composite. The superior tensile properties are also shown by the same composite. A similar trend is also observed when nano CaCO3 is used as filler in NBR (Balachandran and Bhagawan 2012).

The mechanism of transport of toluene across the NBR matrix and CNF reinforced NBR composites can be studied by fitting dynamic swelling data in the Eq. (12) (Southern and Thomas 1967):

$$log{Q}_{t}/{Q}_{\infty }=\text{log}k+nlogt$$
(12)

where Qt and\({ Q}_{\infty }\)are the mole % uptake of solvent at time t, and at equilibrium, the constant k gives information on the polymer-solvent interactions and structural characteristics of the composites, and the constant n give information on the type of transport.

The value of n and k are determined from the slope and intercept of the graph of log Qt/Q ∞ versus logt. The value of n = 0.5 shows a Fickian transport, where the penetration of liquid molecules is slower than the rate of polymer chain relaxation (Jose et al. 2017). If the value n = 1 the transport is non-Fickian, where the penetration of liquid molecules is faster than the rate of polymer chain relaxation. If the value of n lies between 0.5 and 1 then the transport is anomalous. Sometimes a ‘Less Fickian’ mode of transport is also possible when the liquid penetration becomes much lower than polymer chain relaxation, i.e., n<0.5. For all the prepared composites, the n values are between 0.5 and 1 suggests a deviation from the normal Fickian transport and can be considered as anomalous. Visakh et al. reported similar values for n when natural rubber is reinforced with cellulosic nanofibers (Visakh et al. 2012a). The anomalous mode of transport arises because of the coupling of Fickian and non-Fickian behavior. The n value is associated with the response time of NBR segments to the swelling stress and the rearrangement of the polymer chains to provide room for the penetration of the solvent molecules. The segmental motion of NBR chains gets restricted by reinforcing the NBR matrix with CNFs. The slow relaxation of the polymer chains in the presence of CNFs leads to an anomalous transport behavior and a marginal increase in the value of n (Abraham et al. 2015). The magnitude of k signifies the polymer-solvent interaction and structural characteristics of the composite. The value of k is found to be less for NBR/CNF composites compared to neat NBR. This also supports the fact that CNFs can effectively restrict the interaction between the polymer and the solvent molecules. Abraham et al. reported a similar trend in k values when styrene-butadiene rubber is reinforced with MWCNTs (Abraham et al. 2015).

Fig. 3
figure 3

a Stress-strain curves of NBR/CNF composites, b Qt vs. t1/2 plot of neat NBR and NBR/CNF composites, (C) Plot showing the swelling index and cross-link density of NBR/CNF. composites, (D) \(log{Q}_{t}/{Q}_{\infty }\) vs. \(logt\) plot of neat NBRand NBR/CNF composites

Full size image

Fractographic studies of NBR/CNF composites

The failure of CNF reinforced NBR composites mainly occur by the breakage of cellulose nanofibers, and the pull-out of fibers from the NBR matrix. Figure 4(A), 4(B), and 4(C) represent the SEM images of the fractured surfaces of NBR/CNF2, NBR/CNF4, and NBR/CNF6 composites under 600x magnification. Figure 4(D), 4(E), and 4(F) represent the SEM images of the fractured surfaces of NBR/CNF2, NBR/CNF4, NBR/CNF6 composites under 3000x magnification. The smooth continuous phase is seen in the NBR/CNF2 composite [Fig. 4(A) and 4(D)], is an evidence of better interfacial compatibility between the fiber and matrix. No direct separation between CNFs and the NBR matrix was observed at lower filler loading. At low filler loading (2, 4 phr), the cellulosic nanofibers were neither debonded nor pulled-off from the matrix owing to the better interfacial adhesion between CNFs and NBR. A similar fractographic structural pattern was observed in the case of NBR/CNC composites (Suryani et al. 2017). As the fiber content increases, the surface became rough and with irregular fracture patterns. This shows the change in the fracture mechanism. The roughness and special structure pattern [Fig. 4(B) and 4(E)] of the tensile fractured surface of the NBR/CNF4 composite shows better fiber-matrix interaction and effective stress transfer. Figure 4(C) shows agglomerates of CNFs, indicate the inhomogeneous distribution of CNFs in NBR at higher loading (6 phr). The magnified image (Fig. 4(F)) of the tensile fractured surface of the NBR/CNF6 composite showed dispersion issue, deformations, tiny protrusions of CNFs, and clear phase separation. The aggregates of CNFs act as nucleating sites for crack propagation, which will ultimately result in inferior mechanical properties (Correia et al. 2017). No such aggregates are visible in the fractographic images of NBR/CNF2 and NBR/CNF4 composites. The fractographic studies of the prepared nanocomposites are in line with the tensile and sorption results of the NBR-CNF composites. The schematic of the possible hydrogen bonding between CNFs and NBR is shown in Fig. 5.This illustration is drawn based on FTIR, XRD and microscopic observations on the interfacial adhesion between CNFs and NBR.

Fig. 4
figure 4

SEM images of tensile fractured samples of NBR/CNF2 (a) & (D), NBR/CNF4 (b) & (e), NBR/CNF6 (c) & (f) composite under two different magnifications 600× and 3000×)

Full size image
Fig. 5
figure 5

 A possible molecular-level view of the hydrogen bonding between the NBR and CNFs

Full size image

Thermogravimetric and DSC analysis of NBR/CNF composites

Table 4 Thermogram data of NBR/CNF composites
Full size table

The thermal analysis data of NBR and CNF reinforced NBR composites are given in Table 4. The TG and DTG curves of the composites are shown in Fig. 6 A & 6B. The initial weight loss at ~ 100 °C is due to the loss of moisture present in the samples. The degradation of the NBR occurs between 350 and 500 °C in all the prepared composites. The reason for the thermal degradation is polymer chain scission and NBR/CNF cross-link breakage. The onset decomposition temperature (Ton) of NBR, NBR/CNF2, NBR/CNF4, and NBR/CNF6 composites are observed to be 375 °C, 401 °C, 393 °C, and 402 °C respectively. About 26 °C increase in Ton was observed when 2 phr CNFs were introduced into the NBR matrix. Abdul Rashid et al. reported a 5 °C rise in Ton when 2 phr cellulose nanocrystals (CNCs) was used as filler in NBR (Abdul Rashid et al. 2018). The well-dispersed CNFs act as a heat barrier and channelize the heat dissipation in the matrix which in turn prevents the NBR chain scission. The immobilization of the macromolecular chains by the cellulosic nanofibers was probably responsible for the improved onset degradation temperature of the prepared nanocomposites. The temperature at which 50 % decomposition occurs (T50) for NBR recorded an increase from 450 to 461 °C by the introduction of 2 phr of CNFs. The interfacial interaction between the filler and polymer reduces the segmental mobility of the polymer phase in the proximity of CNFs when heating is introduced. Besides, the hydrogen bonding between CNFs and NBR increases the cross-link-density and rigidity of the system which in turn results in higher thermal stability (Abdul Rashid et al. 2018). The interfacial interaction between the CNFs and the NBR matrix protects the fiber by keep away from the direct exposure with the temperature which in turn results in higher thermal stability (Asim et al. 2020).

From the derivative thermogram curve (DTG), the temperature at which maximum degradation occurs (Tmax) of prepared composites were recorded. The Tmax of NBR, NBR/CNF2, NBR/CNF4, and NBR/CNF6 composites are found to be 449 °C, 454 °C, 461 °C, and 459 °C, respectively. About 12 °C rise in Tmax is observed when 4 phr of CNFs is introduced as filler in NBR. Abdul Rashid et al. reported that the Tmax of NBR can be improved from 451 to 465 °C by the introduction of 2 phr CNCs (Abdul Rashid et al. 2018). The increase in Tmax of the prepared composites is mainly due to the restricted thermal motion of NBR chains that are bound to the surface of CNFs (Sreenath et al. 2017). The residue of the prepared nanocomposites at 500 °C was found to be higher at higher CNF loading. The high residue shows the presence of crystalline cellulose in CNFs, which is intrinsically flame resistant. The increase in the residue content reflects the successful incorporation of CNFs in the NBR matrix and better thermal stability.

The kinetic parameters of the thermal decomposition of CNF reinforced NBR nanocomposites are evaluated using the Coats and Redfern method (Coats and Redfern 1964).

The activation energy can be calculated using the Eq. (13):

$$\text{ln}\left(\frac{-\text{ln}\left(1-\alpha \right)}{{T}^{2}}\right)=\text{ln}\left(\frac{AR}{\beta {E}_{a}}\right)-\frac{{E}_{a}}{RT}$$
(13)

where, \(T\) is the heating temperature, A is the Arrhenius- factor, \({E}_{a}\) is the activation energy, R is the gas constant, and \(\beta\) the heating rate. The parameter \(\alpha\) is represented as Eq. (14):

$$\alpha =\frac{{C}_{0}-C}{{C}_{0}-{C}_{f}}$$
(14)

.

Here \({C}_{0}\), \(C\) and \({C}_{f}\) are the mass of the material at the initial temperature, temperature chosen and final temperature, respectively. The plot of \(\text{ln}\left(\frac{-\text{ln}\left(1-\alpha \right)}{{T}^{2}}\right)\)vs. \(1/T\) for neat NBR and NBR/CNF composites is given in Fig. 6 C. The regression analysis gives the slope, intercept, and concurrency value (R2) of the thermal degradation process of NBR/CNF composites. The value of Ea can be calculated from the slope of the curve. The activation energies of NBR, NBR/CNF2, NBR/CNF4, and NBR/CNF6 are found to be 183.64 kJ/mol, 164.91 kJ/mol, 161.58 kJ/mol, and 175.44 kJ/mol respectively. Lower Ea values are observed for NBR-CNF composites when compared to neat NBR. The kinetic study of the thermal decomposition of the prepared nanocomposites showed that the incorporation of CNFs in the NBR matrix accelerates thermal degradation. This is somewhat paradoxical to the result obtained for Ton, T50, and Tmax values of NBR/CNF composites. These results are in agreement with Abdul Rashid et al. (Abdul Rashid et al. 2018). They have reported that Ea of pristine NBR decreased from 88.13 kJ/mol to 56.17 kJ/mol with the introduction of 2 phr CNCs, whereas Tmax was improved from 451 to 465 °C. The possible explanation of the above observation is as follows. The char formed by the decomposition of CNFs may act as a protective barrier to both mass and energy transfer which in turn results in delaying thermal degradation of the NBR-CNF composites. Once the applied heat exceeds a threshold value, the barrier effect of CNFs became ineffective and the plenty of oxygen present in the backbone of CNFs facilitates the thermal degradation of NBR, which leads to a decrease in the activation energy. Li et al. reported a similar observation when porous rice bran carbon was used as a filler in NBR (Li et al. 2014).

The DSC profile of NBR/CNF composites are shown in Fig. 6D. The glass transition temperature (Tg) of the prepared NBR/CNF composites depends upon its structure and segmental mobility of polymer chains. The Tg values of NBR/CNF2, NBR/CNF4, NBR/CNF6 composites are found to be −21.9 °C, −21.7 °C, −22.3 °C respectively. The DSC analysis shows that no significant change in Tg is observed among the prepared composites.

Fig. 6
figure 6

(a) & (b) TG and DTG curves of NBR/CNF composites (c) kinetics of thermal decomposition of NBR/CNF composites using Coats-Redfern equation (d) DSC curves of NBR/CNF composites.

Full size image

Dynamic mechanical analysis

Table 5 DMA data of NBR/CNF composites (NBR, previously published values (Dominic et al. 2020b))
Full size table

To study the viscoelastic properties and constrained region, the dynamic mechanical analysis (DMA) was conducted for the prepared composites. The DMA data are summarized in Table 5. The dynamic mechanical properties of the prepared composites were mainly influenced by the nature of the matrix, type of the filler, the concentration of the filler, and the interface. Figure 7 shows the change in storage modulus (E′) (Fig. 7A), loss modulus (E”) (Fig. 7B), and loss tangent (tanδ) (Fig. 7 C) of the prepared NBR/CNF composites as a function of temperature. The E′ is a parameter to determine the maximum energy stored in the prepared nanocomposites during one cycle of oscillation (Formela et al. 2016). It is a measure of the stiffness and elastic properties of the prepared nanocomposites. The E’ of NBR, NBR/CNF2, NBR/CNF4, and NBR/CNF6 composites at −40 °C (glassy region) is found to be 1189 MPa, 1041 MPa, 1970 MPa and 1833 MPa respectively. The analysis showed that the glassy E′ of NBR/CNF4, NBR/CNF6 composites is found to be greater than that of neat NBR. Surprisingly, in the glassy region, the storage modulus of the NBR/CNF2 composite was found to be less than that of pristine NBR. Liu et al. reported similar observations in the glassy E′ when 1 % chitin nanocrystals were used as a filler in natural rubber (NR) (Liu et al. 2018). They inferred that the slight decrease in the glassy E′ of NR-chitin nanocrystal composite at lower loadings of nanofiller (1 %) was probably due to the change in the conformation and free volume of the rubber. The immobilization of NBR molecular chains by CNFs was probably the reason for the higher E’ of NBR/CNF4 and NBR/CNF6 composites. A sharp transition from glassy to the rubbery state was observed at around −20 °C for all the prepared composites because of the glass transition process. At temperature higher than Tg, the E′ drops and later a plateau is achieved. The rubbery modulus at 25 °C for NBR, NBR/CNF2, NBR/CNF4, and NBR/CNF6 composites is observed to be 3.65 MPa, 6.04 MPa, 8.26 MPa, and 7.18 MPa respectively. Around 126 % increase in rubbery modulus is observed for NBR/CNF4 composite compared to neat NBR. Mariano et al. reported that the rubbery modulus is greatly influenced by the crystallinity index of the nanofiller (Mariano et al. 2016). In other words, the presence of crystalline domains of CNFs substantially improve the rubbery modulus of NBR. Dominic et al. also reported an improvement (68 %) in the rubbery modulus of NBR, when chitin nanowhiskers (2 phr) were used as reinforcing filler (Dominic et al. 2020).

Loss modulus (\(E"\)) is the energy lost by the NBR/CNF composites during the DMA analysis. The increase in the peak height of\(E"\) of the nanocomposites is due to the heat dissipation caused by the filler-filler interaction, filler-polymer interaction during DMA test (Prasertsri and Rattanasom 2012). The Tg was determined from the maximum of E” versus temperature curve. The Tg of NBR, NBR/CNF2, NBR/CNF4, and NBR/CNF6 was recorded as −11.15 °C, −9.51°C, −9.76 °C, and −10.05°C respectively. The positive shift in the Tg of the nanocomposites showed the reinforcing effect of CNFs. The hydrogen bonding between NBR and CNFs at the interface restricts the free movement of the rubber macromolecular chains resulting in an increase in the Tg values of the nanocomposites (Cao et al. 2013b).

Loss tangent (tan δ) is the ratio of E” and E′. The value of tan δ depends on the mobility of the polymer chains (Chen et al. 2015). The tan δ peak height of NBR, NBR/CNF2, NBR/CNF4, and NBR/CNF6 are observed to be 1.535, 0.998, 1.071, and 1.100 respectively. The higher tan δ peak corresponds to a high chain mobility (Zhao et al. 2019). The decrease in tan δ peak height of NBR/CNF composites compared to pristine NBR is probably due to the hindering of the movement of rubber molecular chains by the well-dispersed CNFs (Joseph et al. 2010).The high degree of dispersion of nanofiber results in mechanical interlocking with the NBR matrix which in turn results in intertwining of macromolecular rubber chains. The good dispersion and interaction of CNFs with the NBR matrix decrease the “active” bulk rubber during the dynamic transition process resulting in the reduction in tan δ peak height (Sreenath et al. 2020). The tan δ values showed that the cellulosic nanofibers play a crucial role in decreasing the damping of the prepared nanocomposites. Ziegel & Romanov related the maximum tan δ of the filled and unfilled system in accordance with the Eq. (15) (Ziegel and Romanov 1973)

$$\tan\delta_{mf}=\tan\delta _{mu}/1+\beta\varnothing$$
(15)

Where, \({\tan}{\delta }_{mf}\) and \({\tan}{\delta }_{mu}\) represent the maximum tan δ value of the NBR composites and neat NBR respectively, \(\beta\)is the interaction parameter determining the interfacial interaction between the nanofiber and NBR matrix, \(\phi\)is the volume fraction of nanofiber. The \(\beta \phi\)values of NBR/CNF2, NBR/CNF4, NBR/CNF6 composites were found to be 0.538, 0.433, and 0.395 respectively. The combination parameter ‘\(\beta \phi\)’ value reported in this study is superior to the \(\beta \phi\) value reported for the NBR-precipitated silica system (Boonbumrung et al. 2016). The high value of \(\beta \phi\) at 2 and 4 phr CNF loading showed higher reinforcing efficiency and better interfacial compatibility of CNFs in the NBR matrix.

The Tg with respect to maximum tan δ of pure NBR, NBR/CNF2, NBR/CNF4, and NBR/CNF6 composites is recorded as −4.09 °C, 2.10 °C, 2.67 °C, and 2.13 °C respectively. The Tg of NBR increased by about 6 °C with the introduction of 2 phr CNF. The positive shift in the Tg of the prepared NBR-CNF composites showed the increased cross-link density and matrix-filler interaction (Dileep et al. 2020). Ambilkar et al. reported a similar positive shift in the Tg in the case of NBR- zirconia composites (Ambilkar et al. 2020). To study more about the volume fraction of the polymer chains immobilized, the constrained region (Cv) and adhesion parameter (A) of the CNF reinforced NBR composites were evaluated. The Cv values of NBR/CNF2, NBR/CNF4, and NBR/CNF6 composites are observed to be 0.0845, 0.0693, and 0.0636 respectively. The high Cv values of the composites (2 and 4 phr) is owing to their better cross-link density and good interfacial interaction. The CNFs can effectively restrict the segmental motion of the polymer chain because of hydrogen bonding interaction and nano-size effect. At higher CNF loading, the fibers became aggregated and the filler-filler interactions over ways filler-polymer interactions. That might be responsible for the low Cv value of NBR-CNF6 composite. The adhesion factor ‘A’ was found to be the lowest for the NBR-CNF2 composite followed by 4 phr composite. The lower the value of A, the higher will be the interfacial adhesion between CNFs and the NBR matrix (Dominic et al. 2020).

Fig. 7
figure 7

a Plot of storage modulus versus temperature, b Plot of loss modulus versus temperature, and c Plot of loss tangent (tan δ) versus temperature

Full size image

Conclusions

The manufacturing of value-added products from agricultural waste materials is an emerging area of green engineering. Here, we demonstrated that the cellulose nanofibers extracted from the parasitic plant Cuscuta reflexa could be used as potent green filler in NBR for high-performance elastomeric applications. The FTIR and XRD analysis of NBR/CNF composites ascertained the interfacial interaction between CNF and NBR. The cure time and scorch time of NBR/CNF composites were shown to be greater than that of neat NBR. The cure rate index was reduced in the presence of CNFs. The tensile and tear strength of NBR matrix presented an increase of ~ 125 % and ~ 105 % respectively at 4 phr CNF loading. The remarkable improvement in the modulus, hardness, and abrasion resistance index of NBR-CNF4 composite presents the reinforcing action of CNFs. The compression set of the prepared composites increase with an increase in CNF content. The swelling rate in toluene was minimum for NBR/CNF4 composite owing to their higher crosslink density. The mechanism of transport of the solvent molecules was found to be anomalous rather than Fickian. The Ton, T50, and Tmax significantly improved by the addition of CNFs in the NBR matrix. The Ton, T50, and Tmax of NBR showed an increase of 26 °C, 11 °C and 15 °C respectively at 2 phr CNF loading. The kinetic parameters of the thermal degradation of the prepared NBR/CNF nanocomposites were evaluated from the thermogravimetric curves using Coats-Redfern method. Lower activation energy (Ea) values were observed for NBR/CNF composites when compared to neat NBR. The fractographic studies showed effective stress transfer in the NBR/CNF composites at low CNF loading. Agglomeration of CNFs was observed at 6 phr CNF loading which might be the reason for their inferior physico-mechanical properties. From the DSC analysis, the Tg values of NBR/CNF2, NBR/CNF4, NBR/CNF6 composites were found to be −21.9 °C, −21.7 °C, −22.3 °C respectively. The DMA analysis shows that the Tg of neat NBR can be increased by 6 oC by incorporating 4 phr CNFs ascribed to the reinforcing action of CNFs. The decrease in the loss tangent peak height in the NBR/CNF composites proved the confinement of NBR chains by well-dispersed CNFs. Based on the presented results, it can be concluded that the CNFs can be used as sustainable green filler in NBR for elastomeric applications.

Credit authorship contribution statement.

MD C.D.: Conceptualization, Methodology, Visualization, Writing- Original draft. RJ: Supervision. P.M. SB: Supervision. AS.: Investigation. J. P.A.: Data curation. TJ: Data curation. DP: Resources, KF: Writing, Reviewing and Editing, SS: Writing, Reviewing and Editing, JP: Writing, Reviewing and Editing, MRS: Methodology, Writing, Reviewing and Editing.