Investigation on in situ silica dispersed in natural rubber latex matrix combined with spray sputtering technology

2.1 Materials

Natural rubber latex (NRL) (60 wt% solids) was imported from Thailand. Silica was purchased from Qingdao Croatia Ltd Co. (China). Na₂SiO₃ was provided by Tianjin Hengxing Chemical Reagent Manufacturing Ltd Co. (China). HCl (analytical purity) was obtained from Jinan Shenghe Chemical Ltd Co. (China). NH₄Cl was purchased from Tianjin Beichen-Fangzheng Reagent Factory. Zink oxide, stearic acid, accelerator D (diphenyl guanidine), accelerator DM (2,2′-dibenzothiazoledisulfde), and sulfur were all commercially available industrial-grade products. All investigated silica/NR compounds were based on basic formulation according to Table 1.

2.2 Preparation of NRL/silica slurry in three different ways

2.3 Preparation of NRL/silica masterbatches by spray sputtering method

The amount of silica varied from 5 to 15 to replace carbon black in the rubber matrix, as shown in Table 2. For every 5 phr of silica increased, carbon black decreased by 5 phr and Si₆₉ increase by 1 phr.

The NRL/silica slurry was poured into a container tank. The mixture was atomized by a gas-assisted spray gun under the pressure of 0.5–1.5 MPa. The mixture was sprayed onto a self-designed high-temperature mill roller at 150°C with a scraper. Then, the mixture was atomized into small droplets. After contacting the Teflon coating on a high-temperature mill roll, the moisture contained in droplets rapidly evaporated to produce dry masterbatches. At the same time, the masterbatches with a certain thickness could be scraped and placed aside for further processing.

2.4 Preparation of NRL/silica composites

The obtained masterbatches were refined in an internal mixer (Qingdao University of Science and Technology, China). The process parameters of the mixer were set as follows: filling coefficient 0.7, initial temperature 80°C, rotor speed 80 rpm, and mixing time 6 min. The masterbatches, N660, zink oxide, stearic acid, accelerator D, and accelerator DM were added to the internal mixer, respectively. The mixing cycle time was 5 min. Sulfur was finally added into rubber masterbatches via an open mill (SK-168 from Shanghai Double Wing Rubber Machinery Factory). The process parameters of the open mill were set as follows: roller spacing 3 mm and roll temperature 50°C. And the shear ratio of front and rear rolls was: 1:4. At last, the rubber compound was vulcanized by a QLB-400X2 flat vulcanizing machine (Qingdao Yadong Rubber Machinery Factory) after resting for 16 h.

2.5 Characterization

1. Vulcanization characteristics were carried out according to GB/T16584-1996. The conditions of vulcanization were at 150°C/10 MPa × T90 (min).

2. The Payne effect was tested by the RPA2000 Rubber Dynamic Processing Analyzer (Alpha Technologies Co of USA) with a strain of 0.28–50% and a frequency of 60 rpm.

3. The vulcanizate samples were subjected to brittle fracture in liquid nitrogen and subjected to gold-spray treatment and then analyzed by scanning electron microscopy (JEOL, Japan).

4. The dynamic mechanical properties were characterized by the Dynamic Mechanical Analyzer (GABO Company of Germany). The samples were measured in the fixed frequency mode, at an operating frequency of 10 Hz and a heating rate of 2°C per min. The samples were evaluated in the temperature range from –65°C to 65°C.

5. The tensile strength and tear strength were tested by TS2005b type Tensile Testing Machine from U-CAN Technology Co. (China) according to GB/T528-1999.

6. For tensile tests, the sample was dumb-bell-shaped according to GB/T528-2009 Type 1, and for tear tests, GB/T 529-2008 Type B. Shore A hardness was carried out according to GB/T531-1999.

3.1 Mechanical properties

The mechanical properties amount of silica for NR/silica composites with different processes is shown in Figure 2. From the experiment of variable dosage, the tensile strength and tear strength of the vulcanized rubber prepared by in situ method increase first and then decrease with the increase of the dosage which is different from that of the traditional wet mixing method. The threshold value of in situ preparation of silica appears at 10 phr, while the threshold value of the wet mixing method does not appear in the whole range of variables curve. The reason might be that the silica generated by the in situ method has a smaller particle size of filler agglomeration and is dispersed well with appropriate interaction fillers and rubber matrix via spray sputtering method when 10 phr loading. Thus, the interfacial interaction in silica/NRL composites is improved which forming a regular filler network than wet mixing without in situ silica at 10 phr. Especially, the tensile strength and tear strength of NH₄Cl treated vulcanizates are 34.7% and 19.8% higher than those of ordinary silica vulcanizates, respectively. When the dosage exceeds 10 phr, the properties of the vulcanized composites prepared by the in situ method combined with spray sputtering method are decreased because of the enhancement of the filler network effect (Payne effect) formed among nanoparticles and the reduction of reinforcement effect. This phenomenon will be verified in Section 3.2 with further efforts.

Figure 2

The mechanical properties amount of silica for NR/silica composites with different processes: (a) shore A hardness, (b) percentage of breaking elongation, (c) tear strength, and (d) tensile strength.

It can be concluded from Figure 2 that the tensile strength and tear strength of the vulcanizates obtained by in situ method are significantly higher than those of the wet mixing with commercial silica-filled, while the hardness and elongation at break have no obvious change. In addition, the mechanical properties of vulcanized composites treated by the NH₄Cl method are the most ideal among them. Above all, within a certain range, the mechanical properties of vulcanizates increase as the dosage of silica increases. When silica exceeds a certain amount, the mechanical properties of vulcanizate composites prepared by spray sputtering technology combined with in situ method decrease with the increase of silica content, which reduces the reinforcing effect of fillers on the compound. The reason might be that the silica particles produced in situ have a smaller size of aggregates and uniform distribution, which increases the contact area between rubber hydrocarbons and silica particles. The good adsorption and bonding effect between fillers and rubber composites make the silica-filled in situ combined with spray sputtering method reinforce rubber well and improve the mechanical properties of rubber composites. However, with the increase of silica production, the interaction between filler and filler is greater than that between rubber and filler, and agglomeration occurs to a certain extent which affects the dispersion of silica, thus decreasing the physical and mechanical properties of NRL/silica composites.

3.2 Payne effect of NR/silica composites with different silica dosages

Payne effect (40,41) is a phenomenon in which the shear storage modules G′ of the rubber nonlinear decrease with the increase of strain. Within a certain range, the elastic modulus decreases rapidly with the increase of strain, and the viscoelasticity reaches its maximum value in the region where the elastic modulus decreases. The dynamic modulus of filled rubber decreases sharply with the increase of strain due to the failure of the interaction force between fillers. Therefore, the value of ΔG′ = G′₀ − G′_∞ can indicate the interaction between filler-filler and filler-rubber matrix to some extent. In this section, controlling the same condition via spray sputtering technology combined with the in situ method with NH₄Cl treated, Figure 3 shows the data of 0 phr (ΔG′ = 8.04), 5 phr (ΔG′ = 10.15), 10 phr (ΔG′ = 15.66), and 15 phr (ΔG′ = 22.6) by rubber process analyzer (RPA) test.

Figure 3

The effect of different amounts of silica on Payne effect of NH₄Cl process.

Regularly, the lower the Payne effect is, the better the dispersion of the filler in the rubber matrix and the weaker the network structure of the filler will be. As can be seen from Figure 3, the value of ΔG′ is low with 0 and 5 phr silica which means the interaction force between filler aggregates is not serious, and a less large-scale filler network has been formed. The ΔG′ increased obviously beyond 15 phr silica loaded. As the dosage of silica increases, the distance between fillers and fillers decreases, and the interaction between silica and rubber matrix increases. Therefore, a greater external force is needed to overcome the interaction between silica in the dynamic strain process. However, the degree of filler network is excessive and the reinforcing between rubber and fillers is declined, leading to a lower mechanical property of composites. That might be because the specific surface area of in situ silica is larger, which is more liable to agglomeration. In addition, the interaction between silica and the natural rubber matrix is inherently weak and the developed three-dimensional filler network structure will be decreased under giant dynamic strain. When the filler agglomeration is serious, the filler network has a stronger tendency to collapse under the action of strain, resulting in a larger value of ΔG′. The comprehensive mechanical properties of the compound reach the best with 10 phr silica loaded mainly because 10 phr loading of silica is more compatible with the dispersion capability of this process. Meanwhile, the interfacial interaction between rubber-filler and filler-filler can be beneficial for reinforcing comprehensive properties appropriately.

3.3 Microscopic appearance

Scanning electron microscopes (SEM) tests were conducted on JSM-7500F from JEOL (Japan) with surface gold-spray treatment. In order to compare the particle size of silica generated by different preparation methods and its dispersion in composites, we selected rubber composites filled with different preparation methods under 10 phr dosage for scanning electron microscope observation. When 10 phr of silica is added, the SEM images of rubber composites are shown in Figure 4.

Figure 4

SEM images of composite preparation by spray sputtering method with a different type of silica: (a) silica commercially available, (b) in situ preparation from HCl solution, and (c) in situ preparation from NH₄Cl solution.

It can be concluded from Figure 4a that silica unevenly disperses in vulcanizates obtained by preparation of silica commercially and the local agglomeration is obvious, which decreases the reinforcing effect of rubber. In Figure 4b and c, silica disperses better than the conventional wet mixing method and the aggregate size of fillers is relatively smaller, which verifies the advancement of reinforcement for rubber composites by in situ combined with a spray sputtering method. In case of traditional drying, it is possible for well-dispersed silica to again flocculate which leads to agglomerates. During in situ combined with spray sputtering method, the droplets enfolded in situ silica with appropriate spraying pressure hit the high-temperature roller and then sputter into smaller-grade droplets onto the surface of the roller with instantaneous evaporation of water, which can efficiently improve the dispersion and further enhance the interfacial bonding strength of in situ silica for rubber matrix. However, some aggregates of silica can also be observed by in situ methods from HCl in Figure 4b. That might be because the HCl solution belongs to strong acid and reacts more intensely with Na₂SiO₃ than NH₄Cl with aggregates generation, which leads to worse mechanical properties than in situ preparation from NH₄Cl by spray sputtering process.

3.4 Dynamic mechanical performance analysis

Dynamic mechanics analyzer (DMA) can measure the response of vibration deformation and temperature function of a given material under the action of cyclic stress-strain. Controlling 10 phr silica loading in NR-carbon black matrix, Figure 5 shows the results of loss factor-temperature of NR/silica composites with different processes from −65°C to 65°C. To some degree, the loss factor-temperature curve can reflect filler dispersion, the strength of filler network, wet traction, and rolling resistance of composites with fillers.

Figure 5

Loss factor-temperature of NR/silica composites with different processes (subfigures (b–d) are partially enlarged drawings of subfigure (a)).

In the tire industry, it is difficult to coordinate the relationship among rolling resistance, wet-skid resistance, and wear resistance of traditional tires. These three indexes have become important indexes to evaluate the performance of automobile tires in the modern tire industry. Especially, the loss factor (tan δ) values at two important temperatures of 0°C and 60°C can measure the wet slip resistance and rolling resistance of tires to a certain extent. The value of tan δ near 0°C is used to indicate the anti-skid performance of the tread compound. The larger the tan δ value is, the stronger the holding ability of the tread compound is. The rolling resistance of tread rubber is characterized by the value of tan δ between 40°C and 60°C; the lower value of tan δ indicates the lower rolling resistance and the less heat generated by the friction of the tire on the road surface.

In the glass transition zone, the internal friction of the material mainly comes from the internal friction among the rubber macromolecular chains. The conventional silica-filled method has the lowest peak value which indicates that more rubber molecular chains are trapped in the filler network structure. As a result, a small number of rubber molecular chains participate in the chain segment relaxation process. In other words, the dispersion of fillers in the rubber matrix and the interaction between filler and rubber interface will affect the value of the loss factor tan δ. The peak value of loss factor (tan δ) of in situ silica from the NH₄Cl method is the largest among them because the silica produced by the in situ method has a large contact area and good compatibility with rubber latex. Thus, the agglomerate effect is difficult to achieve, and the strength of the network of composite fillers is weak. There are relatively few rubber molecules retained in the filler network structure, which means that more rubber molecules are involved in the chain segment movement to increase tan δ.

With the increase in temperature, the activity of the rubber molecular chain is enhanced. When the temperature reaches about 60°C, the breaking and reconstruction of the filler network structure in the process of dynamic mechanics is the main reason for the energy loss of the rubber system. The uniform dispersion of silica and the strong adhesion between filler and natural rubber matrix can restrict the movement of rubber molecular chains. Therefore, the energy consumption of breaking and rebuilding the filler network structure is reduced, which leads to the low value of loss factor and nice rolling resistance.