3.1. Curing Characteristics
The curing characteristics of the mixed compounds are given in Table 2
. It is evident that the minimum torque ML
significantly decreased with an increase in the primary carbon black particle size. This is caused by the decreasing amount of the immobilized rubber chains on the decreasing carbon black surface, as well as the structure. The low aggregate structure and surface area of the N 990 type leads to weak interaction with rubber chains. Moreover, the thermal production process leads to a higher purity of the carbon black surface, with low content of active groups. This fact can result in a lower stiffening effect. However, in the case of the torque ML
, interactions between rubber chains and carbon black are mainly caused by physical forces, because the property is in an uncured state [25
The difference between the maximum (MH
) and the minimum (ML
) torques is marked as ΔM
, which is a parameter demonstrating the degree of chemical crosslinking. The same amount and type of curing system was observed in the compounds; thus, the degree of ΔM
is supposed to be comparable for all compounds. In reality, the ΔM
is significantly different. Evidently, the reaction between the rubber and the curing system is one of the factors affecting the crosslinking process [28
The other factor affecting the crosslinking process is the chemical reaction of the rubber with functional groups on the carbon black surface, which differs depending on the carbon black grade. It was found that these functional groups could have either positive or negative effects on the curing characteristics, depending on carbon black type [29
]. Although the N 330 type has a higher surface area compared to the N 550 type, its curing level is lower due to the presence of cure-retarding groups [30
]. This phenomenon causes various t90
values for the compounds in this study.
3.2. Quasi-Static Test Results
The parameters of mechanical properties of the SBR and carbon black compounds are presented in Table 3
. The hardness testing is the most obvious mechanical test. In this case, the Shore A hardness was measured. The obtained hardness results decreased from 69 to 55 Shore A and corresponded to the carbon black particle size, ranging from the largest specific surface (N 110) to the smallest one (N 990). The higher specific surface led to a harder rubber compound. It is evident from the above results that the hardness generally decreased with a decrease in the carbon black surface area.
The tensile test was performed in order to characterize the basic mechanical properties of the rubber compounds. It should be noted that the compounds containing the carbon black with small primary particles (N 110 and N 330) showed the highest break stress (about 23 MPa), while the elongation was the lowest. The increasing primary carbon black particle size causes the decrease of the break stress. This is caused by the reinforcing ability of the carbon black particle size. The smaller primary particle sizes led to a higher specific surface area and to a stronger restriction of the elastomer chain mobility.
Here, the 300% modulus is the parameter characterizing the stiffness of rubber vulcanizates, representing the stress at 300% extension. In addition to previous parameters, it is influenced by the carbon black surface area. The highest value for this modulus was obtained for the N 110 type. The 300% modulus decreased with a decrease in the surface area as well.
The rebound resilience increased from 38% up to 51% for N 110 and N 990 carbon black types, respectively. Additionally, this property is influenced by the particle size, and thus, by the surface area of the carbon black. The stiffening effect of the N 110 type is stronger in comparison to the others. It provides harder rubber with lower rebound elasticity, because a larger part of the mechanical energy is transformed into heat.
Theoretically, the mechanical properties should show greater differences between the N 110 and N 330 carbon black types. Unfortunately, mechanical properties are strongly influenced by the level of filler dispersion in the matrix, which also depends on the carbon black particle size. With decreasing particle size, more energy is required to achieve high filler dispersion. Thus, a shorter mixing time leads to poorer filler dispersion. For this reason, the mechanical properties are reduced. On the other hand, a longer mixing time can cause polymer chain scission and also a decrease in the final properties. For this study, the rubber compounds were prepared under constant mixing conditions.
3.3. Dynamical Mechanical Properties
The effects of various carbon blacks in the SBR matrix on the temperature dependence of the elastic (storage) modulus is summarized in Figure 3
. These curves describe the temperature region, where the hard and brittle material behavior is replaced by the rubbery or viscoelastic behavior. The modulus decreased strongly in the glassy transition region. The elastomer chain mobility, and thereby the glass temperature, is noticeably affected by various additives in the rubber compound. The presence of filler, the amount, an increase in the surface area, or the structure can restrict the chain’s movement ability and lead to the shift of the glassy region to higher or lower temperatures [9
During heating, the elastic modulus of the studied compounds decreased in the temperature range from −50 °C to −25 °C. The most rapid decrease of the elastic modulus was observed for the compound containing the N 990 carbon black type. On the other hand, the highest values of the elastic modulus were obtained for the compounds with the N 110 and N 330 carbon black types. This is connected with the decreasing carbon black particle sizes and the increasing surface area.
The glass transition temperature for the rubber compounds determined from the loss factor (tan δ
) curve of three measurements was in the low temperature region, namely −31.9 ± 0.6 °C, −32.5 ± 0.5 °C, −32.2 ± 0.6 °C, and 31.1 ± 0.8 °C for the N 110, N 330, N 550, and N 990 types, respectively (see Figure 4
). The differences in the glass temperature between compounds were less than 10%. The shift in glass transition temperature was probably caused by sample thickness inhomogeneity and the slightly different clamping forces of each compound. As a result, no significant effect was observed for the particle size on the glass transition temperature. In addition, the damping properties of the rubber compounds can be evaluated from this figure. It is evident that the loss factor increased with an increase in the primary particle size.
The dynamic stiffness of the tested rubber mixtures is measured by the complex modulus of elasticity. The frequency dependencies of the storage modulus and the loss factor at 25 °C are demonstrated in Figure 5
and Figure 6
Significant changes of the storage modulus depending on the frequency for SBR compounds with the N 110, N 330, and N 550 carbon blacks are evident in Figure 5
. There is a moderate increase of the storage modulus with the frequency up to 180 Hz. The filler with the smallest primary particle size (N 110) exhibited the highest storage modulus values, while N 990 gave the lowest values. This is connected with the filler–polymer interaction. This fact depends on the filler’s primary particle size, as well as its structure [28
It was found that the loss factor tan δ
at low frequencies up to 100 Hz showed similar behavior for all carbon black compounds (see Figure 6
). Generally, the loss factor increased with an increase in the frequency. While for the more reinforced carbon black type (N 110) the increase of tan δ
was slower, in the case of the N 990 type the loss factor increased more significantly.
The next measurement was performed in order to evaluate the deformation dependence of the tested samples on their dynamic properties. The elastic modulus dependence on the displacement is shown in Figure 7
. The effects of the carbon black particle size and structure were measured by this experiment. The compound with the smallest primary particle size gave the highest elastic modulus value, while that with the largest particle sizes showed almost no stiffening effect and had almost constant elastic modulus during the whole deformation range. High elastic moduli at low deformation levels are caused by the filler–filler interaction, as explained by Payne [31
]. The smallest particles with higher carbon black structures created stronger filler–filler network in comparison with larger particles and a lower carbon black structures. With the increasing deformation, the filler network is destroyed and the interaction between the rubber and carbon black becomes the main factor.
Incorporated carbon black particles, which are already known as aggregates, create a filler–filler network. Inside the rubber compound, carbon black aggregates are formed by van der Waals forces into so called agglomerates. If a small deformation is applied, a high elastic modulus value is obtained.
The reason for this phenomenon is the strong interaction between filler particles that are not broken. From Payne’s point of view, some of the rubber is immobilized on the filler surface, and in addition some of the rubber is also immobilized inside the branched structure of the agglomerate (known as occluded rubber). If the deformation increases, the agglomerates are broken into smaller sizes, and therefore the elastic modulus decreases. This phenomenon is caused by more mobile smaller units inside the rubber compound. At high deformation levels a plateau can be reached, whereby individual mobile aggregates units are set into motion. This behavior is known as the Payne effect [31
The dependence of the loss modulus on the sample deformation is presented in Figure 8
. The loss modulus increased with a decrease in the primary particle size. The stronger filler–filler interaction led to higher energy dissipation. With the increasing deformation, the loss modulus maximum was found around the displacement amplitude of 35 µm for the compounds with the N 110 and N 330 carbon black types. The highest damping properties were obtained in this area. Apparently, the values of the loss modulus were low for the N 550 and N 990 carbon black types, while damping properties were especially negligible for the N 990 type.
Shear amplitude deformation was measured using a rubber process analyzer. This test describes similar behavior of the rubber compounds as the DMA measurements, but in shear deformation mode.
The dependence of the filler–filler interaction for compounds with the N 110, N 330, N 550, and N 990 carbon black types is depicted in Figure 9
. A strong dependence of the shear storage modulus on the filler particle size is visible, similarly to the DMA testing. Large carbon black particles (N 990) are characterized by a low reinforcing effect, while small particles (N 110) cause a pronounced increase of the storage modulus. Figure 10
shows the strain amplitude dependence of the loss modulus of four carbon black grades. Similarly to the DMA measurements, the shear loss modulus was highest for the N 110 type, while for the N 990 type it was the lowest. The maximum value of the loss modulus was observed at 1% strain, which is fully connected with the maximal damping properties.
3.4. The Hysteresis Characteristics
Due to the viscoelastic nature of the rubber vulcanizates, the stress–strain curves of the tested rubber create a hysteresis loop during the loading and unloading cycles. The hysteresis loop area corresponds to the energy dissipated into heat. The heat generation inside the rubber mixture can lead to it softening and even rupturing. The heat generation is affected by the polymer nature, curing level, and compound composition. This behavior is also known as the Mullins effect.
The hysteresis loops for the studied compounds are presented in Figure 11
. The rubber was dynamically loaded in compression mode during 100 loading cycles. In this figure, the last cycle is recorded. Evidently, there are quite large differences among the carbon black types added to each compound. The largest area of the hysteresis loop was achieved for the N 110 carbon black type, which had the highest specific surface area, while the lowest heat generation was obtained for the N 990 type. An explanation of this phenomena was given by Fukahori [36
]. According to this theory, rubber covers the carbon black surface in creation of the so-called bound rubber. This is an immobilized part of the rubber macromolecules that is physically connected with carbon black particles. During the loading of a polymer–carbon black structure, the orientation of this structure appears. If the unloading process is applied, the stress decreases faster compared to common macromolecular stress relaxation, and the decrease in the unloading curve is visible.
3.5. Vibration Damping Properties
Examples of the frequency dependencies of the transfer damping function of the tested rubber composites measuring t
= 10 mm in thickness with different carbon black particle sizes are shown in Figure 12
. It is evident from this comparison that the size of the carbon black particles has a significant influence on the vibration damping properties. It can be concluded that the material´s ability to damp mechanical vibration generally increased with an increase in the carbon black particle size. This is caused by lower stiffness (or by higher damping) of the rubber composites, which were produced with larger carbon black particle sizes. These facts result in a higher transformation of input mechanical energy into heat during forced oscillations [37
] and a decrease in the values of the damped and undamped natural frequencies [38
]. Therefore, the first resonance frequency (fR1
) value was shifted to the left (see Figure 12
) with an increase in the carbon black particle size, i.e., from 1520 (N 110) to 968 Hz (N 990), as indicated in Table 4
. These findings are in excellent agreement with the results that were experimentally determined by the abovementioned methods, namely the hardness, tensile, shear, and viscoelastic measurements. It was verified in these cases that the increasing carbon black particle size led to a decrease of the break stress, the Shore A hardness, and the storage moduli E´
. In contrast, the rebound resilience was higher for these particle sizes.
The vibration damping properties of the investigated harmonically loaded composite rubber samples are also influenced by their thickness t
, the excitation frequency f
, and the inertial mass m
. The effect of the inertial mass on the vibration damping properties of sample N 330 is shown in Figure 13
. It is visible that better damping properties were obtained with higher inertial mass m
, which led to a decrease in the undamped natural frequency, and thus, the damped frequency. This is due to the fact that the natural frequency of an undamped system is proportional to the square root of the material stiffness for the applied inertial mass [22
]. For this reason, the inertial mass has a positive influence on vibration damping, which is reflected by a shift of the first resonance frequency peak position to lower frequencies, i.e., by the decrease of the fR1
(see Table 4
) from 1441 (m
= 0 g) to 412 Hz (m
= 500 g). The vibration isolation properties of the investigated rubber composites are also significantly influenced by their thickness t
, as shown in Figure 14
for the N 330 sample with an inertial mass m
of 90 g. It is evident that the higher material thickness led to lower values of the fR1
, i.e., from 1138 (t
= 10 mm) to 520 Hz (t
= 20 mm), as indicated in Table 4
. For this reason, the rubber thickness generally has a positive influence on vibration damping properties. It is also visible from Figure 12
, Figure 13
and Figure 14
that the material´s ability to damp mechanical vibration is significantly influenced by the excitation frequency f
. It is evident that resonant mechanical vibration (D
< 0) was achieved at low excitation frequencies, depending on the rubber sample type, the thickness t
, and the inertial mass m
. For example, for the N 110 sample type measuring t
= 10 mm in thickness and without inertial mass (m
= 0 g), the resonant mechanical vibration was observed at frequencies f
< 2950 Hz (see Figure 12
). For the N 990 sample type measuring t
= 20 mm in thickness and with inertial mass m
= 500 g, the resonant mechanical vibration was achieved at considerably lower excitation frequencies (at f
< 330 Hz). In contrast, damped mechanical vibration (D
> 0) was generally achieved at higher excitation frequencies (see Figure 12
, Figure 13
and Figure 14
The material´s ability to damp mechanical vibration was also evaluated for the tested material samples, which were harmonically loaded by the compression force with an amplitude of 10 kN at an excitation frequency of 20 Hz. In the case of the N 990 sample type, it was not possible to perform this evaluation due to the low stiffness of this rubber sample compared to the other tested rubber sample types. The frequency dependencies of the transfer damping function of the investigated rubber samples (thickness t
= 20 mm, inertial mass m
= 90 g) after 750,000 loading cycles are demonstrated in Figure 15
. Again, as in the case of the cyclically unloaded rubber samples (see Figure 12
), the material´s ability to dampen mechanical vibration increased with an increase in the carbon black particle size. For this reason the rubber composites, which were produced with larger carbon black particle sizes, exhibited lower stiffness, resulting in a decrease of the first resonance frequency peak position to lower excitation frequencies. As shown in Table 5
, similar results were obtained independently of the number of loading cycles.
The effect of the number of loading cycles on the vibration damping properties of the N 330 sample type is shown in Figure 16
. It is evident that the vibration damping ability of the sample generally increased with an increase in the number of loading cycles, which led to a decrease of the first resonance frequency peak position to lower frequency values (see Table 5
) with the increasing number of loading cycles. Therefore, the higher number of loading cycles led to a reduction in rubber sample stiffness, which was accompanied by better damping properties in this rubber sample. As shown in Table 5
, similar findings were observed for the other tested rubber composites.