Effects of Filler Functionalization on Filler-Embedded Natural Rubber/Ethylene-Propylene-Diene Monomer Composites

Natural rubber (NR) presents a number of advantages over other types of rubber but has poor resistance to chemicals and aging. The incorporation of ethylene propylene diene monomer (EPDM) into the NR matrix may be able to address this issue. Mineral fillers, such as carbon black (CB) and silica are routinely incorporated into various elastomers owing to their low cost, enhanced processability, good functionality, and high resistance to chemicals and aging. Other fillers have been examined as potential alternatives to CB and silica. In this study, phlogopite was surface-modified using 10 phr of compatibilizers, such as aminopropyltriethoxysilane (A1S), aminoethylaminopropyltrimethoxysilane (A2S), or 3-glycidoxypropyltrimethoxysilane (ES), and mixed with NR/EPDM blends. The effects of untreated and surface-treated phlogopite on the mechanical properties of the rubber blend were then compared with those of common fillers (CB and silica) for rubbers. The incorporation of surface-modified phlogopite into NR/EPDM considerably enhanced various properties. The functionalization of the phlogopite surface using silane-based matters (amino- and epoxide-functionalized) led to excellent compatibility between the rubber matrix and phlogopite, thereby improving diverse properties of the elastomeric composites, with effects analogous to those of CB. The tensile strength and elongation at break of the phlogopite-embedded NR/EPDM composite were lower than those of the CB-incorporated NR/EPDM composite by 30% and 10%, respectively. Among the prepared samples, the ES-functionalized phlogopite showed the best compatibility with the rubber matrix, exhibiting a tensile strength and modulus of composites that were 35% and 18% higher, respectively, compared with those of the untreated phlogopite-incorporated NR/EPDM composite. The ES-functionalized phlogopite/NR/EPDM showed similar strength and higher modulus (by 18%) to the CB/NR/EPDM rubber composite, despite slightly lower elongation at break and toughness. The results of rebound resilience and compression set tests indicated that the elasticity of the surface-modified phlogopite/NR/EPDM rubber composite was higher than that of the silica- and CB-reinforced composites. These improvements could be attributed to enhancements in the physical and chemical interactions among the rubber matrix, stearic acid, and functionalized (compatibilized) phlogopite. Therefore, the functionalized phlogopite can be utilized in a wide range of applications for rubber compounding.


Surface Modification of Phlogopite Using Silane-Based Coupling Agents
Phlogopite was surface-modified with amine or epoxide groups using silane-based coupling agents prior to its incorporation into the rubber blend. The pH of the DI water was set to 3.0 using acetic acid. Two phr of each silane-based coupling agent was added to the modification solution (400 mL), which was subsequently stirred at 300 rpm for 15 min. Thereafter, 200 g of phlogopite was added to the solution, which was stirred once more at 300 rpm for 1 h. The resulting solution was oven-dried at 140 • C for 24 h. The time required for torque to reach 90% of maximum torque during curing (T 90 ) was determined at 170 • C using a rubber rheometer (DRM-100, Daekyung Engineering Co., Ulsan, Korea). The mean of five measurements was calculated for each sample.

Mooney Viscosity
A Mooney viscometer (DWV-200C, Daekyung Engineering Co., Ulsan, Korea) was utilized to measure the Mooney viscosity of rubber samples at 125 • C for 4 min. The sample was pre-heated at 125 • C for 1 min to stabilize the sample prior to measurements.

Tensile Properties
Uniaxial tensile deformation was carried out according to ISO 37 using a universal testing machine (UTM; DUT-500CM, Daekyung Engineering Co., Ulsan, Korea). The cross-section of the specimen was 15 mm × 4 mm and the gauge length was 40 mm. The specimens were elongated at a crosshead speed of 500 mm/min with a load cell of 5 kN at room temperature. The mean values were calculated based on five specimens.

Hardness
A hardness tester (306 L, Pacific Transducer instruments, Los Angeles, CA, USA) was utilized to measure the Shore A hardness of the rubber blends and composites according to ISO 48. The mean values were determined based on five specimens.

Abrasion Resistance
An abrasion resistance test was conducted using an abrasion tester. A 2.5 cm × 2.5 cm sample was rotated 200 times at 40 rpm in a tester with a diameter of 15 cm. The abrasion resistance index (ARI) was calculated based on Equation (1): where ∆m r , p t , ∆m t , and p r represent the mass loss of the reference compound, density of the test rubber, mass loss of the test rubber, and density of the reference compound, respectively.
2.6. Elastic Properties 2.6.1. Rebound Resilience The rebound elasticity of the rubber blends and composites was measured according to ISO 4662 using a ball rebound tester (H090, UTS International Co., Zhangzhou, China). The specimens were maintained at room temperature for 2 h prior to the measurements. A ball was dropped onto the samples, and its rebound height was measured. The mean of five measurements was determined for each sample.

Compression Set
Compression set tests were performed according to ISO 815. The specimens were placed into a cylindrical mold with dimensions of 12.5 mm (±0.5 mm) × 9 mm and compressed at 125 • C for 24 h. Subsequently, the compressed specimen was removed and allowed to rest until it regained its form. The final dimensions (i.e., height) of the specimens were measured and the mean value was determined based on the four specimens.

Morphology
The morphologies of the NR/EPDM blend and NR/EPDM/phlogopite composites were observed using a scanning electron microscope (SEM; Apro, FEI Co., Hillsboro, OR, USA) at an electron beam voltage of 10.0 kV. The surface fractured during tensile tests was coated with a 5-10 nm-thick gold layer using a sputter coater (Cressington 108 Auto Sputter Coater, Ted Pella Inc., Redding, CA, USA) prior to the SEM measurements.

Results and Discussion
Viscosities of unvulcanized rubbers are routinely measured using the Mooney viscosity test. Figure 2a shows the Mooney viscosities of the pristine NR/EPDM blend and its fillerincorporated composites, with and without different compatibilizers. The Mooney viscosity was reduced by incorporating inorganic fillers into the NR/EPDM and the Mooney viscosity of the NR/EPDM composite comprising silica was similar to that of the pristine rubber blend. The phlogopite-embedded NR/EPDM rubber composite showed the lowest Mooney viscosity because of the platy architecture of phlogopite. Mica-based inorganic fillers including phlogopite with the platy architecture usually show low Mooney viscosities [46,47]. The different types of silane-based additives barely influenced the Mooney viscosities of the rubber composites. Different fillers and compositions of rubber composites exhibited different curing behaviors. The optimal curing time for rubbers is routinely defined as the time (T 90 ) required for the torque to reach 90% of the maximum torque during curing. The incorporation of each filler and the different surface modification hardly influenced the T 90 , as shown in Figure 2b. The surface modifications of phlogopite reduced the curing time of the samples, indicating fast curing. Compared with that of the pristine blend, the T 90 values decreased with increasing amino moiety concentration, whereas that of the epoxide-functionalized phlogopite-embedded sample barely changed. A ball was dropped onto the samples, and its rebound height was measured. The mean of five measurements was determined for each sample.

Compression Set
Compression set tests were performed according to ISO 815. The specimens were placed into a cylindrical mold with dimensions of 12.5 mm (±0.5 mm) × 9 mm and compressed at 125 °C for 24 h. Subsequently, the compressed specimen was removed and allowed to rest until it regained its form. The final dimensions (i.e., height) of the specimens were measured and the mean value was determined based on the four specimens.

Morphology
The morphologies of the NR/EPDM blend and NR/EPDM/phlogopite composites were observed using a scanning electron microscope (SEM; Apro, FEI Co., Hillsboro, OR, USA) at an electron beam voltage of 10.0 kV. The surface fractured during tensile tests was coated with a 5-10 nm-thick gold layer using a sputter coater (Cressington 108 Auto Sputter Coater, Ted Pella Inc., Redding, CA, USA) prior to the SEM measurements.

Results and Discussion
Viscosities of unvulcanized rubbers are routinely measured using the Mooney viscosity test. Figure 2a shows the Mooney viscosities of the pristine NR/EPDM blend and its filler-incorporated composites, with and without different compatibilizers. The Mooney viscosity was reduced by incorporating inorganic fillers into the NR/EPDM and the Mooney viscosity of the NR/EPDM composite comprising silica was similar to that of the pristine rubber blend. The phlogopite-embedded NR/EPDM rubber composite showed the lowest Mooney viscosity because of the platy architecture of phlogopite. Micabased inorganic fillers including phlogopite with the platy architecture usually show low Mooney viscosities [46,47]. The different types of silane-based additives barely influenced the Mooney viscosities of the rubber composites. Different fillers and compositions of rubber composites exhibited different curing behaviors. The optimal curing time for rubbers is routinely defined as the time (T90) required for the torque to reach 90% of the maximum torque during curing. The incorporation of each filler and the different surface modification hardly influenced the T90, as shown in Figure 2b. The surface modifications of phlogopite reduced the curing time of the samples, indicating fast curing. Compared with that of the pristine blend, the T90 values decreased with increasing amino moiety concentration, whereas that of the epoxide-functionalized phlogopite-embedded sample barely changed. Tensile properties of the NR/EPDM blend and composites with different fi CB, and phlogopite) and with different phlogopite functionalization types (a epoxide-functionalization) were investigated, as shown in Figure 3. Among th samples, CB-incorporated NR/EPDM composite showed the highest tensile str MPa). The tensile strength of the untreated phlogopite-incorporated NR/E MPa) was analogous to that of the silica-embedded NR/EPDM composite ( However, the functionalization for phlogopite enhanced the tensile strength o posites. The strength of the amino-functionalized phlogopite-embedded com creased as a function of their amino group content. Especially, the epoxide-fun tion for phlogopite substantially improved the tensile strength (14.6 M NR/EPDM composite consisting of epoxide (ES)-functionalized phlogopite, similar to that of CB-infiltrated NR/EPDM composite. The trend of the tensile the composites was similar to that of their tensile strength. The pristine NR/EP without filler showed the lowest modulus. The modulus of the phlogopite-in composites was enhanced by phlogopite surface treatment. The modulus of th tionalized phlogopite-embedded composite was superior to that of the CB composite, indicating that the surface functionalization of phlogopite led to the of a physical network between the fillers and rubber matrix. The elongation silica-embedded NR/EPDM composite was analogous to that of the pristine blend without filler, whereas the elongation at break of the other NR/EPDM was reduced by the incorporation of CB or phlogopite into the rubber blend, of functionalization. Although the ES-functionalized phlogopite-reinforced composite showed the highest modulus, the CB-infiltrated composite showed toughness (energy to break), which was primarily caused by large reduction in at break for phlogopite-incorporated NR/EPDM composites. Tensile properties of the NR/EPDM blend and composites with different fillers (silica, CB, and phlogopite) and with different phlogopite functionalization types (aminoand epoxide-functionalization) were investigated, as shown in Figure 3. Among the prepared samples, CB-incorporated NR/EPDM composite showed the highest tensile strength (14.9 MPa). The tensile strength of the untreated phlogopite-incorporated NR/EPDM (11.2 MPa) was analogous to that of the silica-embedded NR/EPDM composite (11.0 MPa). However, the functionalization for phlogopite enhanced the tensile strength of the composites. The strength of the amino-functionalized phlogopite-embedded composites increased as a function of their amino group content. Especially, the epoxide-functionalization for phlogopite substantially improved the tensile strength (14.6 MPa) of the NR/EPDM composite consisting of epoxide (ES)-functionalized phlogopite, which was similar to that of CB-infiltrated NR/EPDM composite. The trend of the tensile moduli of the composites was similar to that of their tensile strength. The pristine NR/EPDM blend without filler showed the lowest modulus. The modulus of the phlogopite-incorporated composites was enhanced by phlogopite surface treatment. The modulus of the ES-functionalized phlogopite-embedded composite was superior to that of the CB-embedded composite, indicating that the surface functionalization of phlogopite led to the formation of a physical network between the fillers and rubber matrix. The elongation at break of silica-embedded NR/EPDM composite was analogous to that of the pristine NR/EPDM blend without filler, whereas the elongation at break of the other NR/EPDM composites was reduced by the incorporation of CB or phlogopite into the rubber blend, regardless of functionalization. Although the ES-functionalized phlogopite-reinforced NR/EPDM composite showed the highest modulus, the CB-infiltrated composite showed the highest toughness (energy to break), which was primarily caused by large reduction in elongation at break for phlogopite-incorporated NR/EPDM composites. Resistance can be measured when mechanical indentation or abrasion leads to localized plastic deformation. Hardness is defined as the resistance value obtained from such measurements. The hardness values of the silica-or phlogopite-embedded NR/EPDM composites were slightly higher than those of the pristine NR/EPDM blend, as shown in Figure 4. Unlike their tensile properties, the hardness values of the phlogopite-incorporated composites were barely influenced by the functionalization of the phlogopite surface. These results may be attributed to the low concentration (10 phr) of the fillers.  Mechanical properties, such as tensile properties and hardness, are shown in Figures  3 and 4. In addition, abrasion resistance index (ARI) was measured to examine the durability toward abrasion, as shown in Figure 5. ARI is a crucial parameter influencing the performance of rubber materials, especially when they are applied to tires, which experience repeated abrasion. The ARI values were calculated using Equation (1). The infiltration of fillers (i.e., silica, CB, and phlogopite) into the NR/EPDM rubber blend increased the ARI of the resultant specimens. In particular, the CB-reinforced composite exhibited the highest ARI value, compared to other filler-embedded rubber composites. The ARI of the rubber composites rarely showed any change, regardless of the phlogopite functionalization. Resistance can be measured when mechanical indentation or abrasion leads to localized plastic deformation. Hardness is defined as the resistance value obtained from such measurements. The hardness values of the silica-or phlogopite-embedded NR/EPDM composites were slightly higher than those of the pristine NR/EPDM blend, as shown in Figure 4. Unlike their tensile properties, the hardness values of the phlogopite-incorporated composites were barely influenced by the functionalization of the phlogopite surface. These results may be attributed to the low concentration (10 phr) of the fillers. Resistance can be measured when mechanical indentation or abrasion leads to localized plastic deformation. Hardness is defined as the resistance value obtained from such measurements. The hardness values of the silica-or phlogopite-embedded NR/EPDM composites were slightly higher than those of the pristine NR/EPDM blend, as shown in Figure 4. Unlike their tensile properties, the hardness values of the phlogopite-incorporated composites were barely influenced by the functionalization of the phlogopite surface. These results may be attributed to the low concentration (10 phr) of the fillers.  Mechanical properties, such as tensile properties and hardness, are shown in Figures  3 and 4. In addition, abrasion resistance index (ARI) was measured to examine the durability toward abrasion, as shown in Figure 5. ARI is a crucial parameter influencing the performance of rubber materials, especially when they are applied to tires, which experience repeated abrasion. The ARI values were calculated using Equation (1). The infiltration of fillers (i.e., silica, CB, and phlogopite) into the NR/EPDM rubber blend increased the ARI of the resultant specimens. In particular, the CB-reinforced composite exhibited the highest ARI value, compared to other filler-embedded rubber composites. The ARI of the rubber composites rarely showed any change, regardless of the phlogopite functionalization. Mechanical properties, such as tensile properties and hardness, are shown in Figures 3 and 4. In addition, abrasion resistance index (ARI) was measured to examine the durability toward abrasion, as shown in Figure 5. ARI is a crucial parameter influencing the performance of rubber materials, especially when they are applied to tires, which experience repeated abrasion. The ARI values were calculated using Equation (1). The infiltration of fillers (i.e., silica, CB, and phlogopite) into the NR/EPDM rubber blend increased the ARI of the resultant specimens. In particular, the CB-reinforced composite exhibited the highest ARI value, compared to other filler-embedded rubber composites. The ARI of the rubber composites rarely showed any change, regardless of the phlogopite functionalization. The elastic characteristics of rubbers can be investigated by measuring resilience and compression set, as shown in Figure 6. The rebound resilienc ratio of the energy returned to the energy applied for deformation by an ex The compression set represents the degree of deformation sustained when removed after a rubber sample is deformed by a force at a high temperature period of time. Thus, low compression set values indicate high elasticity. The tion of A2S-and ES-functionalized phlogopite into the NR/EPDM rubber matr the rebound resilience and reduced the compression set, which indicates goo However, the elasticity of the A1S-functionalized phlogopite/NR/EPDM wa that of the untreated phlogopite/NR/EPDM composite owing to the low conc amino moieties in A1S, which resulted in low compatibility [48][49][50][51]. The reb ence values of the various phlogopite-embedded composites were higher th the silica-or CB-incorporated composites, whereas compression set values pite/NR/EPDM composites were higher than those of silica-or CB-embedded This finding indicates that the elastic properties for the rubber composites co ica or CB under thermo-pressure conditions during compression set tests we to those of phlogopite-incorporated NR/EPDM rubber composites. Figure 7 shows SEM micrographs of fractured NR/EPDM without and wi phr). The pristine NR/EPDM blend presented smooth surfaces, as shown in F was inhomogeneously dispersed in the NR/EPDM rubber blend, exhibiting tion ( Figure 7b) owing to its extremely fine dimensions. The surfaces of un A1S-treated phlogopite were smooth (Figure 7d,e), whereas some parts of the trix were adsorbed on the surfaces of A2S-and ES-treated phlogopite as show 7f,g. The adsorbed rubber on the phlogopite (red circles in Figure 7) indicat compatibility between these phases, thereby increasing the interfacial adhesi chanical properties. Figure 8 shows the mechanisms of physical and chemical among the rubber matrix, stearic acid, and functionalized phlogopite. The p chemical interactions were caused by A1S/A2S and ES, respectively. The sec The elastic characteristics of rubbers can be investigated by measuring its rebound resilience and compression set, as shown in Figure 6. The rebound resilience refers to a ratio of the energy returned to the energy applied for deformation by an external force. The compression set represents the degree of deformation sustained when the force is removed after a rubber sample is deformed by a force at a high temperature for a certain period of time. Thus, low compression set values indicate high elasticity. The incorporation of A2S-and ES-functionalized phlogopite into the NR/EPDM rubber matrix enhanced the rebound resilience and reduced the compression set, which indicates good elasticity. However, the elasticity of the A1S-functionalized phlogopite/NR/EPDM was similar to that of the untreated phlogopite/NR/EPDM composite owing to the low concentration of amino moieties in A1S, which resulted in low compatibility [48][49][50][51]. The rebound resilience values of the various phlogopite-embedded composites were higher than those of the silicaor CB-incorporated composites, whereas compression set values of phlogopite/NR/EPDM composites were higher than those of silica-or CB-embedded composites. This finding indicates that the elastic properties for the rubber composites containing silica or CB under thermo-pressure conditions during compression set tests were superior to those of phlogopite-incorporated NR/EPDM rubber composites. Figure 7 shows SEM micrographs of fractured NR/EPDM without and with fillers (10 phr). The pristine NR/EPDM blend presented smooth surfaces, as shown in Figure 7a. CB was inhomogeneously dispersed in the NR/EPDM rubber blend, exhibiting agglomeration (Figure 7b) owing to its extremely fine dimensions. The surfaces of untreated and A1S-treated phlogopite were smooth (Figure 7d,e), whereas some parts of the rubber matrix were adsorbed on the surfaces of A2S-and ES-treated phlogopite as shown in Figure 7f,g. The adsorbed rubber on the phlogopite (red circles in Figure 7) indicates excellent compatibility between these phases, thereby increasing the interfacial adhesion and mechanical properties. Figure 8 shows the mechanisms of physical and chemical interactions among the rubber matrix, stearic acid, and functionalized phlogopite. The physical and chemical interactions were caused by A1S/A2S and ES, respectively. The secondary hydroxyl moieties generated during reactions between carboxylic acid and epoxide [52][53][54][55] may contribute additional physical interactions (e.g., hydrogen bonding and dipole-dipole interactions) between hydroxyl and carboxylic groups [56,57].