Functionalized Nanosilica for Vulcanization Efficiency and Mechanical Properties of Natural Rubber Composites

Accelerator functional character was introduced on nanosilica by chemical reaction of sodium isopropyl xanthate (SIPX) with nanosilica (NS). Functional characteristics of nanosilica were confirmed by elemental analysis, thermogravimetric analysis, and infrared spectroscopy. This SIPX functionalized nanosilica (SIPX-NS) incorporated natural rubber (NR) composites were used to evaluate the dispersion of silica in rubber and also the interaction between rubber and filler. The finely dispersed SIPX-NS particles in the NR matrix are revealed from the morphological analysis. Subtle changes in the surface chemistry of silica had a profound influence on dispersibility in the NR matrix. NR 4SIPX-NS composite showed an increase in tensile strength by 10%, flex crack initiation resistance by 13%, tensile strength retention by 16% and cure time reduced by 2 min relative to those of NR 3NS composite. This simple, efficient and cost-effective surface modification of silica improved the vulcanization efficiency and mechanical performance of NR composites and has great potential in the fabrication of high-performance polymer composites.


Introduction
Nanosilica is an emerging environment-friendly nanofiller used for the reinforcement of elastomers and is not derived from petroleum resources [1,2]. These nanoparticles have high surface energy and a tendency to agglomerate because of the groups like siloxane and silanol present on silica surfaces [3]. To overcome this disadvantage, modification of its surface chemistry is necessary to improve its dispersion properties and compatibility in an organic matrix.
The surface modification of silica particles can be achieved by different methods such as organoalkoxysilane modification [4,5], polymer grafting [6,7], encapsulation [8,9], etc. The surface modified silica with organosilanes is used in polymer and rubber industry for the improving mechanical properties. The surface interface between the polymer and filler surface affects the nonlinear viscoelastic behaviour and improves the mechanical properties as well as energy efficiency of rubber nanocomposites [10]. Zhang et al. found that the addition of a silane coupling agent (Si 69) modified silica reduced the rolling resistance and heat build-up of NR composites [11]. Chen et al. proposed a surface modification of silica with sulphur monochloride by the reaction between silanol hydroxyl groups in silica and chlorine atom in sulphur monochloride [12]. This silica-supported sulphur monochloride improved the cure efficiency and tensile strength of SBR composites. The silica-supported vulcanizing accelerator was prepared by grafting 2-benzothiazolethiol onto the surface of silane modified silica to improve the silica-matrix interaction and silica dispersion in SBR [13]. Mathew et al. found that SBR composites containing plasma treated silica could improve the tensile strength and modulus of SBR composites [14]. Wang et al. developed surface modified silica nanoparticles by combining noncovalent and covalent modification processes in a simple, efficient and cost-effective method [15]. Weng et al. also developed a new protocol to promote the dispersion of silica and interfacial strength along with reduced abrasion loss in SBR/silica composites by grafting with oniums [16]. Liu et al. established a simple inhibition-grafting method to prepare silica/polydimethylsiloxane nanocomposites with superior tensile strength, tear strength and low viscosity [17]. Guo et al. directly blended sorbic acid during SBR/silica compounding and the addition of 15 phr sorbic acid increased the storage modulus and reduced the glass transition temperature due to the deagglomeration of silica particles in SBR matrix [18]. Gill et al. prepared NBR/Chitosan/ nanosilanized silica blends and observed improved crosslink density and hardness upon nano-silanized silica loading [19].
Natural rubber, an unsaturated elastomer has been widely used due to its excellent elastic property. However, low polarity, poor oil resistance and low air impermeability of NR limit its applications in some cases [20]. To overcome these drawbacks, NR is modified appropriately. Xu et al. used epoxidised natural rubber (ENR) as an interfacial modifier to improve the mechanical and dynamical mechanical properties of NR/silica composites [21]. The results indicate the formation of a covalent bond by the ring-opening reaction between the epoxy groups of ENR chains and Si-OH groups on the silica surfaces. Ismail et al. selected maleated natural rubber as a coupling agent for paper sludge fiber filled NR composites and found excellent rheological and dynamic properties [22]. According to Gelling, NR backbone stereo-regularity was disrupted by any type of chemical modifications and hence, reduction in basic strength properties of NR [23].
After a thorough investigation of the literature, we came to know that the surface modification of silica with suitable coupling agents is a classical way to improve the interfacial adhesion between particles of silica and natural rubber. The surface modification not only improves the hydrophilic-hydrophobic interactions between silica and NR but also enhances the dispersion of silica in the NR matrix. Even though the silane modification has several advantages, the high cost of silane coupling agents and the high processing temperature required for the efficient surface modification limit its applications. So, the surface modification of silica with a costeffective and efficient surface modifier with unique properties is a need of the hour. Herein, we report the use of sodium isopropyl xanthate (SIPX) as a novel surface modifier for silica. To our best knowledge, no studies have been reported so far to evaluate the role of SIPX modified silica in NR. Nanosilica with high surface area (520 m2/g) derived from bamboo leaves in the present study is also remarkable. The surface modification and its size distribution were characterized by FTIR, TGA, EDX, SEM and dynamic light scattering (DLS). Modification of nanosilica improved curing, mechanical performance, crosslink density and aging resistance of NR.

Modification of Nanosilica with Vulcanizing Accelerator SIPX
5 g of SIPX was dissolved in 50 mL of isopropyl alcohol and to this solution, 10 g of nanosilica (dried at 100°C for 6 h) was added. This mixture was heated at 80°C for 24 h with continuous stirring. The unreacted SIPX was then removed from the reaction mixture by soxhlet extraction using isopropyl alcohol for about 8 h. The residue obtained was vacuum dried at 50°C for 3 h to obtain SIPX modified nanosilica (SIPX-NS).

Composites Preparation
Thermo Haake Polylab with a rotor speed of 60 rpm maintained at 70°C was used for compounding as per ASTM D 3184. Table 1 gives the details of formulation. Initially, mastication of NR was done for 3 min. Stearic acid, zinc oxide and styrenated phenol were then added followed by DEG and nanosilica and the compound was mixed for 3 min. TMTD, CBS and sulphur were added and the mixing was continued for 2 min. After 8 min of mixing, the compound was sheeted out at 5 mm nip gap (5 times) by using a laboratory size (6"×12") two-roll mill and finally sheeted out the compound at a 3 mm nip gap. Before moulding, this compound was allowed to mature for 24 h at room temperature. A hydraulic press having 12"×12" platen size was used to vulcanize this compound to the optimum cure time at 150°C temperature and 150 kg/cm 2 pressure.

Fourier Transform Infrared Spectroscopy
Fourier transform infrared analysis was conducted using Thermo Nicolet, Avatar 370 model IR spectrometer, in 4000-400 cm −1 spectral range with a resolution of 4 cm −1 .

Energy Dispersive X-Ray Spectroscopy
JEOL (JED-2300 Model) instrument was used to perform EDX (Energy Dispersive X-ray) analysis. Malvern mastersizer-V3.30 instrument capable of measuring size between 0.1 to 1000 μm and an angular range of 0.032-60 degrees was used to measure the particle size distribution.

Dynamic Light Scattering
The particle size of nanosilica dispersion in water was measured using Horiba scientific nano particle analyzer SZ-100 with a scattering angle of 173 degrees.

Scanning Electron Microscope
The scanning electron microscope of JOEL (Model JSM 8390 LV) was used to analyze the tensile fractured surfaces. Sample surfaces were sputtered with a thin layer of gold before the analysis to avoid charging on the surface.

Thermogravimetric Analysis
Thermogravimetric analyser (TA instruments, model Q-50) was used to perform thermogravimetric analysis. Samples were kept in a nitrogen atmosphere and heated from room temperature to 750°C at a heating rate of 20°C/min.

Thermal Conductivity
Thermal conductivity was measured according to ASTM D7340 using Holmarc's Lee's Disc Apparatus (Model: HOAE-LD18). The apparatus comprises a brass disc resting on another slab of the same dimension with a special heating coil.

Tensile and Tear Test
Universal Testing Machine from Instron, USA was used to analyze the stress-strain properties of samples as per ASTM D 412. Test specimen un-nicked at 90°angle was used for the tear resistance test according to ASTM D 624. ASTM D 573 method was used to carry out the thermal aging analysis.
The following equation was used to determine the tensile retention percentage: Tensile retention% ¼ Tensile strengh after aging Tensile strength before aging X100 ð1Þ

Hardness and Abrasion Resistance
The shore A hardness of compression moulded samples was measured as per ASTM D 2240 using Bariess durometer. The tests were performed on unstressed samples of 6 mm thickness. A load of 12.5 N was applied and the readings were taken after 10 s of indentation when a firm contact has been established with the specimen. The abrasion resistance of the samples was determined using Bariess DIN abrader, Germany (ASTM D 5963). The sample having a diameter of 16 ± 0.2 mm was kept on a rotating sample holder and a 10 N load was applied. Initially, a pre-run was given to the sample and its weight taken. The weight after the final run was also noted.

Specific Gravity and Compression Set
Specific gravity was determined as per ASTM D 297 using Densimeter at room temperature. ASTM D 395 standard aws used to measure the compression set. The samples (12.5 mm thick and 29 mm diameter) in duplicate compressed to a

Heat Build-Up and Flex Crack
Heat build-up study was carried out using Dynesco Goodrich Flexometer according to ASTM D 623. The test pieces were prepared in a cylindrical shape with a diameter of 17.8 ± 0.1 mm and 25 ± 0.15 mm height by the compression moulding machine. Demattia flexing machine was used to check flex cracking and crack growth of the samples as per ASTM D 430 and ASTM D 813 respectively.

Rebound Resilience
Rebound resilience was measured according to ASTM D 7121 using Wallace Dunlop Tripsometer.
The following equation was used to calculate rebound resilience percentage (RB)

Crosslink Density, Swelling Index and Mol Percentage Uptake
Crosslink density, swelling index and mol percentage uptake were measured by the equilibrium swelling method. Flory-Rehner equation was used to calculate the crosslink density of samples [25].
Where M c is the molar mass of the sample between consecutive crosslinks and this can be calculated using eq. 4.
In eq. 4, V s is the molar volume of solvent (for toluene,106.2 cm 3 /mol); ρ r is the rubber density (0.94 g/ cm 3 ); V r is the rubber volume fraction of samples at equilibrium swelling and χ is the interaction parameter between natural rubber and toluene (0.3787 from the literature [26]). Ellis and Welding equation [27] was used to calculate V r For all samples, eq. 6 was used to calculate the percentage solvent uptake (Qt%) Qt The following equation was used to determine the swelling index.
Here, Ws and Wi are the swollen and initial weight of the specimen. Figure 1 shows the FTIR spectra of samples NS, SIPX and SIPX-NS. In NS, the silanol hydroxyl groups stretching vibration and adsorbed moisture together constitute the peak at 3560 cm −1 . The peaks at 807 cm −1 and 1107 cm −1 are due to the symmetric and asymmetric stretching vibrations of Si-O-Si linkages [17]. The bending vibrations of -OH groups are responsible for a peak at 1642 cm −1 . In SIPX, the C=S stretching vibration leads to the absorption band at 1050 cm −1 and 1037 cm −1 [28]. The peak at 1191 cm −1 is due to C-O-C symmetric stretching vibration [29]. The peaks at 1377 cm −1 corresponds to bagging of the C-H bond in the CH 3 group of SIPX [30]. The C=S vibration peak of SIPX-NS at 1037 cm −1 shifts towards lower wavenumbers 1002 cm −1 and the peak at 1050 cm −1 shifts towards higher wavenumbers 1118 cm −1 compared with those of SIPX. The shift in these peaks is due to the coordinated structure of SIPX-NS.

Elemental Analysis
The elemental analysis of the SIPX modified nanosilica shows carbon, sulphur and sodium besides silicon and oxygen. The percentage of elements is tabulated in Table 2. The results confirm the successful modification of the silica with SIPX.

Dynamic Light Scattering Analysis (DLS)
The particle size distribution of SIPX-NS was measured using DLS. As shown in fig. 2(a), the average particle size for NS was 6 nm. The particle size of SIPX-NS was increased to 20 nm and this may be due to the modification and mild agglomeration of NS particles. Figure 3 shows the scanning electron micrographs of NS and SIPX-NS particles. SIPX-NS particles show morphological changes and a bigger size compared to NS. The average size of SIPX-NS measured from the micrograph was 20 nm, whereas for NS the particle size was 6 nm. This result is in agreement with the results obtained from DLS studies. Hydrogen bonding and Van der Waals forces among SIPX-NS particles lead to agglomeration also [31].

Thermogravimetric Analysis
Thermograms of NS and SIPX-NS are shown in fig. 4. Within the temperature range of 30-750°C, two weight loss steps were exhibited by nanosilica. The release of adsorbed  Tensile stress-strain curves obtained for NR composites moisture is the reason for the initial weight loss on heating from 50 to 110°C [32]. Dehydration of silanol groups occur between 400 and 600°C results in the second stage weight loss [33]. The decomposition of modifying group in the temperature range of 150-600°C is the reason for the major weight loss for SIPX-NS. The very low initial weight loss of SIPX-NS between 50 and 110°C is due to the increase in hydrophobic character of silica by the presence of an organic modifying group. Analysis of percentage residue confirmed the presence of 9.3 wt% of SIPX on silica particles. Figure 5a illustrates the cure characteristics of natural rubber composites. Table 3 provides the cure parameters. The surface property, concentration and nature of filler influence the cure behavior of composites [34]. Compared to NR SIPX-NS composites, the scorch and cure time of NR gum and NR 3NS composites are higher. The increase in the cure time of NR 3NS composite may be due to the adsorption of curatives on the particles of silica. Silica particle adsorption of curatives leads to an increase in cure time for NR 3NS composite [35]. Lowering of NR SIPX-NS composite optimum cure time (t 90 ) is due to the reduction in surface hydroxyl groups of nanosilica and also the accelerating effect of SIPX present on the modified silica. The scorch time (ts 2 ) of modified nanosilica composites is drastically reduced due to the accelerating efficiency of SIPX. Reduction in scorch time indicates that the NR SIPX-NS composites have lower processibility. The torque values are increased with increasing SIPX-NS concentration due to the better crosslink density of composites, as indicated in Table 3. Figure 5b shows the derivative cure curve of NR composites. It indicates the accelerating effect of modified silica on the curing reaction of NR composites. From the graph, it is clear that NR SIPX-NS composites achieved the maximum peak height quickly during the progress of reaction compared to NR gum and NR 3NS composites. The maximum peak obtained for NR gum and NR 3NS composites is at 3 min range, whereas for NR 3SIPX-NS, NR 4SIPX-NS and NR 5SIPX-NS are only at 1 min range. Figure 6 shows the stress-strain behaviour obtained for all NR composites. Higher tensile strength was obtained for NR 4SIPX-NS compared to all other composites. Better rubber and filler interaction is the reason for higher tensile strength. At a higher concentration of SIPX-NS, aggregation of particles of silica causes poor interaction between the filler and the rubber [36]. In our previous work on natural rubber reinforced nanosilica composites [37], we found that the maximum tensile strength is obtained for the NR 3NS composite. In NR SIPX modified nanosilica composites NR 4SIPX-NS shows a 10% and 24% increase in tensile strength compared to NR 3NS composite and NR gum compound respectively.

Stress-Strain Behaviour of NR Composites
Elongation at break percentage was found to be decreased with an increase in silica concentration. This is because of the restriction in the movement of polymer chains by the presence of non-deformable silica particles [38]. An increase in modified nanosilica concentration leads to an increase in tear strength and modulus at 300% elongation. This is due to higher crosslink density and bound rubber content [11]. Elongation at break, tensile strength, modulus at 300% elongation and tear strength obtained for natural rubber composites are given in Table 4. The concentration of filler, particle size and its dispersion affect the tensile strength and modulus of the composites. The tensile properties of the composites were studied after aging at 100°C for 24 h, as it is very important to evaluate the performance in practical applications. The percentage retention of the tensile strength is shown in Fig. 7. NR gum shows lower retention compared to nanosilica-filled composites. Luo et al. [39] observed a similar result while studying the effect of silica and antioxidant on NR composites. NR 4SIPX-NS shows excellent tensile retention in comparison with all other natural rubber composites. Table 5 includes the tabulated values of other technological properties of NR composites. The composites showed an increase in hardness with an increase of filler loading. This is in line with the increase in modulus and crosslink density [40] of the composites.

Other Technological Properties of NR Composites
The abrasion resistance of rubber vulcanizates depends on several factors such as filler particle size, structure, surface activity, and filler-rubber interaction [41]. The reduction in abrasion loss for NR SIPX-NS composites attributed to the improvement in service life due to the good interaction between the matrix and filler and better filler dispersion.
The ability of a material to recover from an applied continuous strain is measured in compression set analysis. The percentage of dynamic compression set and compression set values found to be increased with an increase in nanosilica concentration as silica is a non-resilient reinforcing filler [37]. With the increase in modulus of the composite, more restrictions in polymer chain mobility occurred even after the applied stress is removed. This is confirmed from modulus values which show a linear relationship with the compression set.
In the heat build-up test, the dissipation of energy occurs as heat because of the friction among filler particles and also between rubber matrix and filler under cyclic deformation [42]. With the increase in filler loading heat build-up of composites found to be increased. This causes a rise in fatigue failure and leads to inferior mechanical properties of the composites [43]. There is no significant change in heat build-up for NR 4SIPX-NS and NR 3NS samples as the modification improve the filler dispersion in the rubber matrix. Table 6 shows the initial flex crack and crack growth of NR silica composites. Resistance to flex crack and crack growth is the essential dynamic properties required for the rubber products used in dynamic applications. Flex crack resistance of the NR composites mainly depends on the dispersion of filler,   [44,45]. Modified nanosilica filled NR composites showed higher resistance to flex cracking and crack growth compared to unmodified nanosilica filled NR composite. This improved resistance may be due to the presence of long hydrocarbon chains in SIPX-NS, which helps better dispersion of nanosilica and improved crosslinking [46]. The lower value for both flex cracking and crack growth resistance for unmodified nanosilica NR composite was due to the agglomeration of nanosilica particles and rigid interaction of nanosilica with NR. The maximum resistance obtained for NR 4SIPX-NS composite was due to the optimum sulphur-accelerator ratio at this concentration [47]. Tear strength and fatigue life are interconnected as both are related to the breaking energy of the composite [48]. The relationship between crack initiation and tear strength is plotted in fig. 8. The plot implies that the factors which contribute to improve tear strength may be the same as the factors which contribute resistance to crack initiation. The crack initiation cycles of NR 5SIPX-NS were lower compared to NR 4SIPX-NS due to the denser crosslink formation [48], as tabulated in Table 3.

Morphology of NR Composites
The dispersion of filler in the polymer matrix determines the ultimate properties of composite samples. Stress concentration points developed inside the composites due to aggregation of particulate fillers which may lead to inferior properties while nanoparticles uniform distribution enhances the properties [11]. A little roughness and few agglomerates of silica present on NR 3NS tensile fractured surface ( fig. 9(a)) indicate that the interfacial interaction between nanosilica and NR is less. NR 4SIPX-NS composite shown in fig. 9(b) is rugged with undulations and uniform filler dispersion, indicating that the matrix could transfer the applied stress to the nanosilica during the tensile test. Figure 10 shows the thermogravimetric curves obtained for the composite samples with nanosilica, modified nanosilica and gum compound. NR gum and composites showed a single step degradation pattern. The initiation of this single step  degradation occurs by the thermal C-C bond chain scission followed by hydrogen transfer at the scission site [49]. NR gum showed T on (degradation temperature at 10% wt. loss) at 325°C. 3 phr nanosilica addition in NR increased T on by 3°C while 4SIPX-NS had no significant change due to the initial decomposition of SIPX on nanosilica particles. The maximum degradation temperature and temperature at 50% degradation (T 50 ) of both NR 4SIPX-NS and NR 3NS composites were same and 4°C higher compared to gum compound. The degradation of NR depends upon the nature of metal ion present, concentration of xanthate used etc. [50].

Thermal Stability of NR Nanosilica Composites
As the amount of xanthate in SIPX-NS is low, the composite with SIPX-NS has no significant effect on thermal properties. The thermal degradation characteristics are shown in Table 7.

Thermal Conductivity of NR Composites
Thermal conductivity data for NR gum, NR 3NS and NR 4SIPX-NS composites are shown in Table 8. The thermal conductivity of polymers is influenced by the incorporation of fillers [51]. The filler nature, orientation, dispersion in the matrix, volume fraction and thermal conductivity of the filler are the major parameters contributing to thermal conductivity of composites [52]. Thermal conductivity values of NR nanosilica composites were found to be slightly higher compared to NR gum as nanosilica has a low value of thermal conductivity [53]. Minor enhancement in thermal conductivity of NR 4SIPX-NS composite compared to NR 3NS composite may be due to the better interaction, which is beneficial for the transmission of a phonon with decreased phonon scattering [54].

Swelling Behaviour and Crosslink Density of NR Silica Composites
Rubber-filler interaction and chemical crosslinks inside the vulcanized composites definitely affect the crosslink density [55]. Solvent molecular size, mobility of polymer chains and free volume inside the composite are the main factors that affect the transport of solvent through the rubber matrix. Table 9 shows the tabulated values of swelling index and crosslink density of composites samples. NR composites showed a decrease in solvent uptake with an increase in silica concentration. Compared to the NR 4SIPX-NS composite NR 3NS sample showed higher toluene uptake. This is due to better silica-rubber interactions and uniform filler dispersion in NR 4SIPX-NS composite. The transportation of solvent molecules through the NR matrix is restricted by the uniformly distributed SIPX-NS particles. After vulcanization, mobility of the NR chains is restricted by filler particles. Maximum crosslink density was observed for the NR 5SIPX-NS sample and the value was comparable with the 4SIPX-NS containing NR composite.

Conclusion
We have studied the synergistic effects of the accelerator sodium isopropyl xanthate and nanosilica on the thermooxidative aging resistance and cure characteristics of natural rubber. We have also developed a simple and highly efficient surface modification of nanosilica using sodium isopropyl xanthate. The modification was confirmed by FTIR, EDX, FESEM, DLS and TGA. Improved rheological, mechanical and aging properties were observed for the modified silica NR composites, which could be credited to the significantly improved rubber-filler interfacial interaction, higher crosslink density and dispersion. Improved interfacial interaction and dispersion of modified silica were revealed by FESEM. NR modified silica composite had high crosslinking density and hence showed high tensile strength retention and flex crack resistance. This indicates the composite has a long service life and can be used for a wide range of applications.
Availability of Data and Material My research didn't generate any data or I reused existing data.