The effect of pumice and clay composition in natural rubber-ethylene propylene diene monomer blends towards its curing characteristics and physic-mechanical properties

Since some adverse effects of using carbon black as petroleum-based filler in rubber compounding, mineral fillers may be a feasible alternative filler. This research studied the effect of the composition of pumice and clay as mineral filler in natural rubber (NR) and ethylene propylene diene monomer (EPDM) blending towards its curing characteristics and physical-mechanical properties. The effect of reinforced filled in EPDM-NR (70/30) phr blends was observed by varying 50, 70, and 90 phr of filler. While the filler was designed as the composition of pumice/clay (P/C): 20/80; 50/50; 80:20 (weight/weight). The rubber, fillers, and chemical were compounded based on ASTM D-3182. The results showed that the loading of P/C fillers in EPDM-NR blends increased scorch time and decreased cure rate index. It was found that the optimum number of filler could enhance density, modulus. Remarkably, the higher pumice composition caused the abrasion resistance and compression set improved significantly, whereas the overloading pumice and clay cause the physic-mechanical properties remained stable or evenly drop. FTIR spectra of the binary blends confirmed that the fine fillers dispersion was obtained due to some interaction bonding between silanol, aluminol of fillers with the EPDM-NR.


Introduction
Natural rubber is a polymer material with high elasticity that is widely used for tyres, rubber pad, rail pad, gasket, seal, rubber dock fender Etc. The advantages of using natural rubber as the main polymer are excellent tensile properties, abrasion resistance and highly availability in nature [1]. However, some negative effects of using natural rubber are less chemical resistance, low physic mechanic properties, and oil resistance for rubber product applications are urgent to be addressed [1,2]. Generally, in order to improve those properties, specific rubber products were made using a ICSTSI 2020 IOP Conf. Series: Materials Science and Engineering 980 (2020) 012003 IOP Publishing doi: 10.1088/1757-899X/980/1/012003 2 combination of synthetic polymers such as ethylene diene, styrene butadiene, chloroprene, or even combined with other thermoplastic polymer such as poly-vinyl chloride, polyethylene low molecular or polypropylene [1][2][3]. In addition, to increase physical properties such as hardness, density, and other tensile properties some fillers were blended during rubber compounding. In this research, EPDM, petroleum-based synthetic polymer, was studied since it was blended with NR to produce some rubber products application that needs high abrasion and heat resistance, and specific compression set value. However, in order to improve physical properties such as density and hardness, the compatible reinforcing needs to be added as a filler in rubber compounding. The previous research stated unfilled EPDM, NR or EPDM-NR binary blends are weak and lost its physical and tensile properties [1,[4][5][6]. EPDM blends with NR have also been studied and stated that to obtain better tensile and mechanical properties, the plasticizer and fillers need to be compatible with each other with the rubber matrix [7,8].
Commercial reinforcing fillers such as carbon black are usually added into the rubber phase matrix to reinforce the polymer's properties of polymer and reduce cost. Some non-reinforcing fillers such as Calcium Carbonates is added in rubber compounding to enlarge the products volume. However, some negative effect have risen due to petroleum based on carbon black such as environmental pollution, high price and low sustainability recently [9]. Thus, some deposited mineral from earth has been studied as either reinforce or non-reinforce filler for rubber compounding. Some previous research using clay from montmorillonite, kaolinite, quartz, etc. to improve physical properties NR blending [9]. It is found that the strengthening effects are determined by its large surface area, porous, size and surface polarity of the fillers [4,6,10]. The type and amount of filler determine the rubber products' physical-mechanical properties, and thermal and aging resistance. Since many deposited minerals have very specific physical and chemical properties, a study on the effect of using some minerals needs toward its physic-mechanical properties is urged. The type, amount and its compatibility of filler with polymers determine the specification and standard of the rubber final products' specification and standard.
In this research, similarly with the zeosil as reported by [3,11], pumice and clay were studied due to the physical and chemical properties of those minerals that may be reinforced filler in rubber compounding. Montmorillonite, bentonite and other clay commonly used as filler in rubber compounding [5,9,12,13]. It is reported that clay has good positive effects on the physical properties of natural rubber composites. Meanwhile, pumice is a new alternative reinforced filler that has not been explored widely. Pumice fine powder is a natural mineral that has high availability from explosive eruptions [14][15][16]. Pookmanee, Wannawek [16] stated that generally pumice in Indonesia has chemical composition as follows 70% SiO2; 14% Al2O3; 4.39% Fe2O3; 4.39 %; 3.77% Na2O; 4.03% K2O; 2.6% CaO, 0.71% TiO2, Etc. This mineral contain also reported by Bahri and Rahmaniar [12]. Naturally, the pumice has a crystalline structure due to alumina, silanol, and hydroxyl groups [15][16][17]. It has a highly porous glass that has huge surface area [15]. The polar groups on the surface can be mixed easily with the polar polymer [17]. Because of its properties, pumice has mostly been used as an adsorbent, poly aniline mixtures, and other industrial application [17]. In addition, pumice also has excellent chemical resistance and thermal stability [16,17]. The clay added in this research was obtained from Banyuasin. The density was 1.6 g/cm 3 that dominated by kaolinite and a small amount of smectite, quartz, crysto-ballite and quartz in a very small portion [18,19]. Thus, the pumice and clay mixtures may have positive effects on physic-mechanical properties that are not had by carbon black, calcium carbonates, kaolin and other commercial fillers.
Bearing in mind that lack of information and research on pumice usage reinforces filler in NR/EPDM blending, this research would be valuable as reinforcing filler in polymer blends. Therefore, based on some excellent properties of pumice, clay and natural rubber properties, this research studied to prepare a pumice-clay based composite of EPDM-NR. This study emphasized on the formulation of rubber compounding particularly the effect of the number of pumice and clay as fillers towards it density, modulus, abrasion resistance, and compression set. Therefore, some information and knowledge on developing reinforced fillers based on mineral may be published.

Pumice and clay preparation
Pumice rock was bought from the local market, Sumatera, Indonesia. The pumice rock was dried at sun light, crushed and ground until obtaining a fine pumice powder. Then, it was sieved at size 200 mesh, dried at the temperature of 70-80ºC using air oven and then desiccated to maintain the moisture content before compounded. The density of pumice used in this research was 0.71 g/cm 3 .
Clay was obtained from a local pottery market in Banyuasin, Sumatera Regency, Indonesia. Similar to the pumice preparation, the clay was dried, ground to be clay powder, and then sieved at 200 mesh size powder. Both of clay and pumice were maintained the moisture content before compounded with polymers and other chemicals.

Rubber compounding
The vulcanizate was compounded based on the formulation as follows: 40 phr of EPDM; 60 phr of natural rubber type: Ribbed Smoke Sheet (RSS-1) ; 2 phr of paraffinic oil, 2phr of brown factice, 4 phr of Zinc Oxide; 1.5 of Stearic Acid; 0.5 of Paraffin Wax; 1.5 of CBS; 0.5 phr of DPG; 2.5 phr of Sulphur. The total number of filler was varied 50; 70; and 70 phr for rubber compounding. Meanwhile, the filler was a mixture of pumice and clay that also varied by mixing pumice/clay (weight/weight) as follows 20/80; 50/50; and 80/20 w/w. Polymer and other chemicals were compounded using laboratory mill XK-160 (type: open two roll mill, Shanghai Rubber, China) based on the standard procedure of ASTM D3182. The preparation of vulcanized sheets standard was done using the ASTM D3184-89 procedure. In order to examine the tensile properties, the compounded rubber sheet was vulcanized for t90 at 150ºC using heating press hydraulic (YG-220, Shanghai Rubber Machine Worker, China). These procedures were replicated three for each formula in order to obtain highly accurate data.

Testing procedures
Curing characteristics such as ML, MH, cure time and scorch time were determined using a Monsanto moving to die rheometer (HT8756P) according to ASTM D5289. Then the samples were tested at a vulcanization temperature of 155ºC.
For physical-mechanical and tensile properties, the compounded rubber was sheeted and cooled for about 24 hours at room temperature. After that, the sample was heated and pressed using hydraulic press machine YG-220, China, for about 20 minutes at 155ºC. The samples were tested based on the standard procedures as follows: modulus (ASTM D412); abrasion resistance (ASTM D5963); Compression set at 25% deflection, 70ºC for 22 hours (ASTM D395); and density (ASTM D297). The experiments were conducted three replication to obtain accurate results.
In order to observe the functional group interaction between clay, pumice, natural rubber, and EPDM, the spectra of FTIR was pictured. The FTIR of Perkin Elmer System 2000 was used in a range 4000 cm -1 to 500 cm -. The evaluation of the spectra was following the ASTM D3677 procedure. The rheology analysis on the effect of filler number and the composition of pumice and clay filler in EPDM and NR blending are shown in table 1. It lists the torque differences that reflect the crosslinking degree of the matrix polymer and also the curing rate index which describes the filler's processing ability of the filler within the polymers. The higher loading of total filler of pumice and clay caused the increase of MH (maximum torque), indicating viscosity due to the high number filler. Similarly, the increased loading of total filler also contributed to the increasing of torque difference (DF = MH -ML) that is assumed as cross-link density. It could be explained that silanol and alumina from both clay and pumice could make interaction among the ethylene and isoprene groups within the rubber matrixes as reported by Ramesan, George [17]. Pumice and clay with a specific area and large porous also could adsorb some chemicals such as sulphur curing agents, plasticizers and chemicals during the rubber compounding [6]. In addition, the high number of pumice and clay also may inhibit the curative agent and other chemicals dispersion within the rubber matrixes as reported Sengupta, Esseghir [10]. The existence of excess sulphur and precursor cross linking also inhibit the curing process. Therefore, it affect the cure time of the rubber blending. In addition both pumice and clay surfaces have a high number of hydroxyl groups and water (as hydrate) that may span the vulcanization of EPDM-NR blending [1,4,6].
As the pumice and clay increased, the CRI relatively decrease. The EPDM and NR have a huge molecular weight that needs some extra energy to melt mixing during the mastication and compounding with the big-sized filler such as pumice and clay [5,13,20]. Hence, the more shear force and heat were needed during melting the mixture due to the high viscosity and the presence of plasticizer and softener in the compounding.
Those higher viscosity of EPDM-NR filled with pumice and clay may shorten the scorch time as listed in table 1. Compared to previous research by [5,13,20], this finding met an agreement that the clay loading on EPDM-NR blending influenced the curing and rheological characteristics. The higher total filler that had a big size of 200 mesh of pumice and clay could also limit the rubber matrixes' molecular movement within the rubber matrixes. Thus, it would affect the physical properties like 300% modulus and density as depicted in figure 1. The pumice and clay molecule restriction could affect the stiffness and rigidity of the rubber products as the density value shown in figure 1.
The value of t90 (scorch time) explained the processing-ability of the filler and polymer of the rubber mixing the mastication and compounding process particularly. This process was connected with the amount of heat that could be energized before the rubber cross link-state transformation. The increasing viscosity of filler may shorten the scorch time as reported by some research by Ismail and Mathialagan [6]. The excess number of filler loading within EPDM-NR blending may cause the decreasing of cure time. The competition for cross-linking agents between NR and EPDM and the overload pumice and clay may cause a decrease of cure time. The competition for cross-linking agents between NR-EPDM and the overload pumice and clay casue the decreasing the rate of curing due to the less mobility of the curative agent in highly dense of EPDM, NR, pumice, and clay.
Process-ability and blending of fillers within the EPDM-NR is affected by the presence of pumice and clay [6,10]. Pumice and clay has hydroxyl groups that may enhance the rubber filler networking due to the hydrogen bonding and van der walls effect. The same explanation was also reported by Azizli, Abbasizadeh [21]. It was shown that the stronger bonding interaction between EPDM, NR and pumice and clay fillers that was determined by ΔF (MH-ML) in table 1. In addition, the pumice and clay dispersion in a sufficient composition of EPDM-NR matrixes also contribute to the cross link density as assumed by the ΔF. The presence of pumice and clay contributed positively to the increasing of density as shown in figure  1, whereas it decreased the modulus of the polymer blending. The density increased as the loading of pumice and clay, but at an optimum number of filler the density tended to remain stable. Conversely, the modulus of the polymers experienced the decreasing by the increasing filler. The overloading of pumice and clay cause the aggregation of the filler and some chemicals may agglomerate due to the presence of excessive filler [5,11,13]. It might interrupt the filler matrix bonding of oligomer of alumina, silica, isoprene from the filler and polymer matrixes. Thus it reduced the modulus properties as shown in figure 1. As confirmed by the cure rate index in table 1, the modulus might be represented as cross-linking density and its stiffness [6]. Compatibility between rubber and the fillers of mixture between pumice and clay influences the rubber compounding's processing ability. The fine filler dispersion may enable the stronger rubber filler interaction that lead to improve the modulus and other tensile properties [4,6]. However, due to the big size of pumice and clay fillers, at a specific portion of excessive fillers, the pumice and clay may agglomerate and its aggregate could decrease the strength of filler rubber interaction and even rubber-rubber networking bonding [5,13,20]. Thus, physically, the rubber products are more fragile and stiff.

Mechanical properties
The effect of pumice and clay loading towards its abrasion resistance and compression set is shown in figure 2. Pumice and clay loading had a less significant effect on abrasion resistance and compression set value as reported by some research [22,23]. Due to the scrapping, the volume loss, mechanical rubbing and erosive action is a key mechanical properties for rubber products such as tire, rail pad, rubber pad, etc. Figure 3 illustrates that filler loading of pumice and clay increase the abrasion resistance of the rubber blending. Unlike the modulus trend, better abrasion resistance could be obtained by improving of the filler rubber interfacial adhesion and rubber-rubber networking bonding. The value of abrasion resistance in figure 2 indicated the loss material during the testing, so the lower value indicated the higher mechanical properties. The pumice and clay addition of 50 up to 90 phr might fill the optimum space on the EPDM-NR composites matrix finely, so that its strong bonding interaction could resist the friction and decrease the lost volume. The homogenous particle size of clay and pumice at 200 mesh also contributed to the strength cross-link density and its hardness so that the abrasive treatment could be resisted [4,6]. As shown in figure 2, due to the differences bulk density of pumice and clay, the abrasion resistance of composites containing high number of pumice had the lowest value of abrasion resistance. This finding is in agreement with some research by Tabsan, Wirasate [22]. Another important mechanical property is the compression set. The compression set's value was measured by evaluate the changes in thickness and elasticity of vulcanization after pressed at a certain deflation and temperature [2,13]. This property contributed to flow, stress relaxation, recover from compression or setting during rubber product usage. The residual deformation or the percentage of deflection after compressed 25% of its original thickness. These properties were very important in rubber padding for heavy construction or transportation. Figure 2 shows the deflection percentage of NR/EPDM filled pumice and clay rubber products that recover to its early shape and how much the rubber failed to return to its previous shape. Generally, the lower percentage indicated a good compression set. It was clearly seen that the compression set value of the rubber less than 30%. It indicated the compression set of this ternary blends complied good the specification standard of rail rubber pad. However, for static usage such as gasket or rubber sealing, the loss of ability to seal may cause leakage, therefore the low compression set is needed for this application. Apart from shrinkage and swelling properties by contacting with the fluids, more prolonged usage and increasing temperature increase the compression set. It was found that the increase of pumice clay total filler caused a positive effect on the improvement of compression set. Similarly to abrasion resistance, the higher number of pumice also gave a lower value of the compression set. The bulk density and chemical properties pumice and clay may enhance the decreasing of mechanical properties of the rubber products due to deterioration in rubber filler interaction bonding and finely filler rubber dispersion [17].
Aforementioned the density and modulus changes by the increasing number of pumice and clay, the compression set also shown a similar trend as shown in figure 2. The formation of more cross-linking bonds, the stronger interaction could resist the changes during compress and set at a certain condition. The previous studies also reported that the number filler and its type of reinforcing   [10,13,20,21]. Those results are in consistent with the 300% modulus in figure 1, which the rubber stiffness was reduced, so the rubber became more elastic and the compression sets improved.

FTIR spectra
FTIR spectra were pictured in figure 3. Compared with EPDM-NR filled carbon black, the spectra of EPDM-NR filled pumice and clay have a higher intensity and more peaks in specific wavenumber. The spectra showed that the higher number of total filler loading might increase the peaks' intensity at specific wavenumber. As depicted in figure 3, adsorption bands derived from mixture clay and pumice were observed as follows: 450-460 cm -1 (Si-O-Al), 750-770 cm -1 ( Si-O-H) , 1010-1300 cm -1 (Si-O-Si) and 3400-3450 cm -1 (OH). Those adsorption bands also explained as the same functional groups [15][16][17]. The intense high peaks approximately at 3450, 3000, 1200, 1000 cm -1 were the vibration of -OH, C-H, C=C, and Si-OH respectively. It was highlighted that pumice and clay were dispersed in EPDM/NR caused new changes of spectra in 1000-4000 cm -1 and some new adsorption bands. This was due to the presence of aluminol, silanol, and the interaction between other functional groups in pumice and clay within the polaric groups of NR and EPDM. It was noticeable that peak intensity at 3450 cm -1 became broader and higher due to the presence of OH from pumice/clay and some hydrogen bonding between silanol and aluminol groups from pumice/clay with isoprene and ethylene propylene. Ramesan, George [17] also state the same clarification on filler dispersion with the rubber matrixes. Thus, the dispersion of pumice/clay within binary blends was obtained.

Conclusions
The binary blending of NR and EPDM filled by pumice and clay was successfully compounded. To sum up, pumice and clay were possible alternative as reinforcing filler in EPDM-NR polymers that