1. Properties Improvement of Nanocellulose-Reinforced Rubber Composites

This section would encompass the mechanical improvements achieved through the incorporation of various isolated NCs in rubber polymers. Mechanical behavior of the rubber, such as tensile strength, elongation at break and Young’s modulus, was mainly emphasized along with strain energy density (SED) from the recent highlighted studies. Briefly, SED or sometimes known as modulus of toughness, is the amount of strain energy that a material can absorb per unit volume before fracture. It is considered as an energetic local way to investigate fatigue failure and fracture in static conditions by making a postulation that brittle fracture happens when the local SED reaches a critical value, known as the critical strain energy ^[1]. SED values can be determined by mathematical integration to obtain the areas under stress–strain curves, and a larger SED would translate to improved material ductility ^[1]. The Young’s modulus (gradient of the stress–strain curve in the linear elastic region), on the other hand, may be very small as most elastic regions are short or, in some cases, non-existent or not well defined, especially in rubber materials ^[2]. Therefore, researchers opt to define “modulus” at a certain strain percentage as a substitute such as M100, M200 and M300. Sample stress–strain curve for a rubber material is shown in Figure 1 with relevant terminologies labeled.

Figure 1. Sample stress–strain curve for rubber materials.

1.1. Nanocellulose-Reinforced Natural Rubber Composites

The use of NC from a myriad of sources as a reinforcing agent in NR or blends is comprehensively studied. The use of NR as the dispersed aqueous phase is a perfect candidate model system to study the improvement effects of nanofiller reinforcement due to its excellent flexibility ^[3].

The incorporation of NC from different sources into the NR matrix has received increasing interests in industries. NC would be isolated from its natural source through chemical or mechanical means prior to being mixed into the latex. There is no requirement of complex machinery nor elevated operating temperature to blend NC into the matrix. Depending on the scale of sample production, the use of roll mills may be necessary. At a glance, the incorporation of NC into NR has yielded a significant improvement in tensile strength across numerous studies, regardless of NC source and type. Most reports compare samples with and without the presence of NC constituents in vulcanized NR (which include the addition of compounding agents such as accelerators and stabilizers). This indicates that the addition of NC is compatible with existing compounding systems and delivers its function as a reinforcing material ^[4][5]. Furthermore, a common trend could be observed where elongation at break values decreased, and sample moduli increased. Elongation at break values are related to the ductility of the material, and its decrement could be caused by the agglomeration of NC particles within the rubber matrix ^[6]. The increase of sample moduli is greatly related to the tensile strength and could be attributed to the restriction of polymer chain movement by the presence of NC particles within the rubber chain network.

Studies from Dominic et al. ^[7] and Kulshrestha et al. ^[8] show that partial substitution of CB by CNF of 5 phr and up to 15 phr, respectively, in NR performed substantially well in terms of mechanical strength when compared to systems fully reinforced by CB alone. This suggests that the use of NC could phase out the conventional filler in years to come. In addition, some investigations also involve the addition of foreign reagents to improve interactions between the nanofiller and the rubber matrix, either through surface modification of NCs or adding dispersants in rubber latex. For instance, Cao et al. ^[9] used carboxylated tunicate CNC in ENR matrices and observed concentration-dependent improvements of crosslink density and an approximate 50% improvement of the tensile strength (from 2.3 MPa to 3.5 MPa) when compared to unmodified tunicate CNC at 5 phr loading. This was attributed to the orientation of modified CNC and rubber chains, which induced stronger interfacial covalent bonds for effective stress transfer at the filler-matrix interfaces. Jiang and Gu ^[4] added resorcinol-hexamethylenetetramine (RH) into CNCs obtained from four different sources and studied its potential to improve filler-matrix interactions. As a prominent gelation reagent, RH provided good adhesion properties between filler and rubber chains and improved the dispersion of CNCs. This resulted in increased tensile strengths of the nanocomposite of 20% compared to samples without RH. In another study, Parambath Kanoth et al. ^[10] used free-radical thiol-ene chemistry to modify CNC as an improved nanofiller in NR. The addition of mercapto groups to the CNC doubled the tensile strength when compared to samples without modification, and approximately 5-fold compared to neat NR. The increment in elongation at break also suggests that this functionalization also provided elasticity to the rubber.

A comprehensive study by Jardin et al. ^[11] focuses on the percolation effects of the presence of CNCs in the matrix of natural as well as synthetic rubber. No CNC surface modification was performed, and mutual dispersibility of CNC and rubber latex in water was exploited through thin sheet sample preparation methodologies. Although there were polarity differences between hydrophilic CNC and hydrophobic rubber, the difference in chain structure made agglomeration issues less potent in NR. This was due to the presence of a more pronounced steric hindrance between the CNC and rubber, which limits intimate interfacial interactions between the filler and host matrix ^[11]. Figure 2a shows specks of CNC (as indicated by the arrows) well-dispersed in the NR latex sheet, whereas Figure 2b shows some percolation of CNC with the formation of CNC networks. Figure 2c,d show that percolation networks are more widely formed with higher concentrations of CNC, which contributes to its greatly improved tensile strength by 8-fold at 6 phr CNC as illustrated in Figure 2e. The formation of percolation effects due to strong filler interactions improves the tear strengths of the sample as it would be more difficult for tearing forces to travel along paths with the least resistance, as shown in Figure 2f. Another investigation by Yu et al. ^[12] using honeycomb-like structured regenerated nanocellulose (RC) explained that the inclusion of NC in rubber matrices contributes to mechanical improvement in many ways. Referring to Figure 3, hydrodynamic effects arising between the filler and NR, strong interactions due to interlacing phenomena as well, as percolation effects contribute to effective stress distribution, resulting in extraordinary improvements of the tensile strength up to 8.5 times and global modulus up to 29-fold of the nanocomposite at 30 phr RC.

Figure 2. (a,b) SEM cross-sectional image of natural rubber (NR)/CNC sheets; (c,d) TEM images showing CNC percolation in formulated rubber thin sheets; (e) representative tensile curves for NR/CNC nanocomposite thin sheets, where “NB“ denotes non-broken samples, and “B“ denotes fully broken sample; (f) proposed schematic of strong filler–filler interparticle forces as an outcome of percolation effects. Adapted with permission from ^[11]. Copyright 2020 Elsevier.

The mechanism of reinforcing NR with the use of an array of NC derivatives is somewhat similar across many studies. In general, the addition of untreated NC into NR latex would reduce mechanical performance due to poor adhesion and agglomeration. The effects of these issues could be minimized or delayed up to higher filler concentrations through surface modification of nanofillers or adding compatibilizers into the rubber latex. However, they could not be eliminated. As previously mentioned, NC derivatives are commonly hydrophilic, whereas rubber matrices are hydrophobic, causing reduced intermolecular interactions between filler and host matrix as well as differences in polarity. Furthermore, when untreated NC compounds are mixed into NR latex, the strain-induced crystallization of NR would be seriously compromised ^[13]. Agglomeration issues are also common when dealing with NC-rubber nanocomposites. Over a threshold limit of NC fillers present in the rubber matrix, a saturation of the filler becomes visible and causes mechanical performance to deteriorate extensively. From the perspective of CNFs, fiber agglomeration results in the formation of large bundles and interaction between fibers outweigh that of between the fiber and the matrix ^[13]. As a result, voids would form and become stress concentration points that cause the nanocomposite material to fail prematurely through facilitated crack propagation ^[11][14]. NC nanofillers are efficacious in strengthening NR nanocomposites through effective stress transfer, which is attributed to their large aspect ratio and exceptional strength independently ^[4]. At optimal nanofiller concentration and excellent dispersion, NC fillers could interact with each other through hydrogen bonding, which also aids in distributing the stress over the polymer matrix ^[13]. When stress is applied to the nanocomposite, it is transferred from the host matrix to the nanofillers, which also contributes to a greater tensile modulus. The common trend of reduction in elongation at break values indicates that adding NC nanofillers into NR latex affects its elasticity and stiffness. This value continues to decrease upon the increasing concentration of nanofiller addition and is attributed to the immobilization of rubber polymer chains by well-dispersed nanofiller networks. Incorporation of NC fillers reduces the mobility of the host matrix, resulting in greater stiffness and lower fracture strain values.

Figure 3. Schematic illustration of NR reinforced with regenerated cellulose in alkaline urea–aqueous system. Reprinted with permission from ^[12]. Copyright 2017 American Chemical Society.

1.2. Nanocellulose-Reinforced Synthetic Rubber Composites

Alike that of NC-reinforced NR composites, synthetic rubber could also be further reinforced with a variety of NC compounds. However, the issues of dispersibility, agglomeration and interfacial bonding remain while the host polymer network varies. Reports have shown that the strain-induced crystallization phenomenon also occurs in synthetic rubber, similar to that of NR, as the phenomenon is mainly controlled by nucleation processes, which are directly proportional to the strain rate ^[15]. This greatly depends on the protruding molecular chains along the polymer backbone and its lower stereoregularity ^[16]. The mechanisms of synthetic rubber reinforcement are comparable to that of its natural counterparts, such as the formation of percolation networks for effective stress transfer through improved interactions between filler molecules as well as between fillers and the host matrix.

There has been an increasing trend towards surface modification of NC fillers prior to addition to the synthetic rubber latex. Synthetic rubber matrices could be selective in their compatibility with surface modifiers, hence opening more opportunities for research and investigations. The application of NC in synthetic rubber is more scarce compared to that in NR as properties of rubber from synthetic origins could be tailored from the formulation stages and may not require additional filler material on top of existing ones. Many studies venture into SBR or IR as the representative synthetic rubber due to its practical significance and a wider range of commercial applications ^[17]. Scientists are aware of the environmental implications of conventional compounding agents and are exploring routes to reduce the carbon footprint of rubber processing. Hence, using natural and sustainable materials such as NC is a promising solution to the issue.

Xu et al. ^[18] studied the application of bagasse NC grafted with maleic anhydride (MAH) and styrene as an improvement to neat NC, with prospects of partial substitution of conventional CB filler in SBR matrices. They found that the grafted fillers proved to be more efficient as a reinforcing material through suppression of the Payne effect while increasing modulus and hardness up to a threshold value. The modified NC increased the tensile strength of SBR to a maximum of 32.5 MPa from 30.0 MPa at 10 phr CNC loading, while neat NC showed negative reinforcement potential mostly attributed to agglomeration.

In another study, Sinclair et al. ^[19] explored the potential of CNFs in SBR matrices in terms of nanofiller loading and functional agents. They reported that a moderate concentration of 7 phr CNF managed to improve the tensile strength of neat SBR to about 8 MPa, almost triple the initial value, with significant improvements to Young’s modulus. It was noted that the improvement was contributed by effective load sharing between the CNF and SBR matrix and the formation of a CNF percolation network for efficient stress load distribution. Conversely, a decrease in mechanical strength is common at higher filler loadings, suggesting aggregation of reinforcement agents and formation of defects within the matrix, as depicted in the SEM images in Figure 4a–h. The pristine CNF was functionalized with a plethora of functional agents such as 3-mercaptopropanoic acid (T3), 11-mercaptoundecanoic acid (T11), 4-pentanoic acid (A4), 10-undecenoic acid (A10) and cysteine (TC). Based on Figure 4j, the use of functionalized CNF managed to mimic the stress–strain curve patterns of industrial SBR, which has high tensile strength and moderate strain. CNFs functionalized with mercapto groups (TC-CNF, T3-CNF, and T11-CNF) yielded comparable results despite their difference in chain lengths. This was similar to vinyl-functionalized CNF (A4-CNF and A10-CNF). When compared between groups, A4-CNF provided more significant improvements to SBR in terms of strength (10.5 MPa against 9.4 MPa) and modulus (12.6 MPa against 9.7 MPa) at 7% CNF concentration. These findings indicate that vinyl groups have greater hydrophilic reduction abilities in CNF, which improved the linkages between CNF and SBR during vulcanization.

Fumagalli et al. ^[20] researched the potential of surface-modified CNCs and CNFs with 3,3′-dithiodipropionic acid chloride (DTACl) in SBR. They reported the occurrence of a reactive interface and strong stiffening behavior with the addition of the modified nanofillers. Impressively, 10 wt% of DTACl-modified CNC and CNF managed to improve the tensile strength of neat SBR by 7-fold and 5-fold, respectively. Other properties such as modulus and elongation at break increased as well, making them comparable to characteristics of rubber reinforced by industrial CB and silica fillers. Furthermore, a study by Wang et al. ^[21] used bacterial cellulose whiskers without modifications in XNBR and found that the reinforcing potential peaked at 13 phr, providing quadruple improvement in terms of tensile strength, while elongation at break decreased slightly. The tear strength of the sample was also doubled. This was attributed to facile stress transfer through H-bonds within the percolating bacterial nanocellulose network.

Figure 4. SEM images of tensile fractured surfaces of neat styrene-butadiene rubber (SBR) and SBR/CNFs nanocomposites. (a,b) Neat SBR; (c,d) 3% CNFs; (e,f) 7% CNFs; and (g,h) 9% CNFs. Representative tensile stress−strain curves of SBR nanocomposites reinforced with (i) pristine CNFs and (j) various functionalized CNFs. Adapted with permission from ^[19]. Copyright 2019 American Chemical Society.

2. Potential Applications of Nanocellulose-Reinforced Rubber Composites

Recently, naturally derived nanomaterials are playing a crucial role in various fields, including wearables, transport, and biomedical science. However, challenges such as ease of purification methods, mass production and their practical applications remain major concerns ^[22]. Despite these challenges, some studies involving NC-reinforced rubber composites managed to show notable results. For instance, Nagatani ^[23] fabricated a sponge-rubber material based on NR/CNF composites for sports shoe sole applications. Crosslinkers, such as dicumyl peroxide, were used to produce the material with azodicarbonamide as a chemical blowing agent. The CNF surface modification through oleoylation had greater reinforcing effects and endowed it with hydrophobic properties. The presence of double bonds on the functionalized side chains of the CNF can form crosslinks by reacting with sulfur in rubber compounds ^[23]. Based on the positive outcome, a sports shoe sole was developed consisting of the composite CNF whose matrix consisted of a blend of ethylene-vinyl acetate copolymer and NR. This robust material is under development to further exploit its benefits and improve the current prototype model of the lightweight CNF-reinforced shoes ^[23].

Similarly, studies by Visakh et al. ^[24] and Abraham et al. ^[14] showed the potential uses of NC-reinforced NR for barrier membrane applications. In the former investigation, the nanocomposites were prepared with crosslinking agents, activators, accelerators, and a set amount of NC dispersion through ball milling and ultrasonication techniques. It was hypothesized that the formation of a zinc-cellulose network complex between ZnO (as an activator) and NC coexisted with the crosslinked NR network. The polarity of the cellulose molecules enabled a strong interconnecting network within the composite structure. As a result, the NC-reinforced NR composite exhibited reduced solvent absorption rates against benzene, toluene, and p-xylene ^[14][24]. The diffusion coefficient also had a decreasing trend against increasing NC concentration. It was proposed that, compared to CNC, the separation efficiency was more efficient when CNF was incorporated into NR due to tangling effects of the nanofibers ^[14]. These studies demonstrate the potential use of NC-reinforced rubber as a membrane barrier material for the separation of organic solvents in addition to enhancing its mechanical integrity.

Another novel application of NC-reinforced rubber is in the field of electronics and wearable sensors. In this context, a report by Silva et al. ^[25] highlighted the potential use of functionalized CNF/PANI and NR nanocomposite materials in terms of mechanical properties and electrical conductivity. Briefly, CNF was coated with PANI through in-situ polymerization prior to incorporation into the NR matrix. Samples from the study showed that the addition of unfunctionalized CNF into NR improved the tensile strength by 4-fold and functionalized CNF/PANI by more than double. This could be explained by the greater hydrophobicity of CNF compared to CNF/PANI, which results in improved adhesion to the NR polymer chains ^[25]. Furthermore, upon testing the samples for their electrical conductivity for wearable sensor applications, it was found that the addition of functionalized CNF/PANI in NR increased conductivity of the material as compared to unfunctionalized CNF by about 10-fold ^[25]. The presence of PANI chains enables the hopping of free charge carriers, which translates to electrical signal conductivity. Thus, the addition of NC into a rubber not only toughens the polymer matrix but could also be functionalized to endow the material with electrical properties for wearable electronics.

Separately, another study by Phomrak et al. ^[26] formulated NR latex foam reinforced with BC and NC. In their study, NC was initially dispersed in water, followed by thorough mixing in NR latex. Potassium oleate soap was added to the mixture to make foam until the volume was tripled. Other compounding agents like accelerators, gelling agents and antioxidants were then added and homogenized. The composite porous foams fabricated with the Dunlop method showed increasing trends of tensile strength up to 15 phr of NC addition. Furthermore, with the aim of the composite material to be a sustainable shock absorber or supporting material, compressibility tests were also conducted. It was highlighted that the addition of NC in the NR latex foam also enhances compression recovery of up to 35% due to enhanced adhesion and molecular interactions between NC and NR. Samples in this study showed insignificant effects on thermal stability regardless of the concentration of NC addition, hence drawing the conclusion that the NC-reinforced foam can be used for applications at high temperatures up to 300 ℃ ^[26]. These studies show the successes of ongoing investigations regarding applications of nanocellulose-reinforced rubber materials. To ensure continued success, close research communications between the industries and research institutions are essential to make these material innovations feasible and affordable globally.