Effects of graphene and carbon nanotubes on mechanical enhancement and heat generation of rubber composites

Carbon nanofillers can enhance both thermal and mechanical properties of rubbers due to their high thermal conductivity and tensile strength at the nanoscale. But the effects of CNTs and GR on heat generation of rubber composites are yet unclear. In this paper, carbon nanotubes/carbon black/nature rubber (CNTs/CB/NR) composites and graphene/carbon black/nature rubber (GR/CB/NR) composites were prepared by using GR and CNTs as partial replacements for carbon black, respectively. The interfacial interactions between fillers and the rubber matrix were calculated by the contact angle test, the filler-matrix network was characterized by rubber processing analyzer, and the glass transition temperature of the composite was measured by dynamic mechanical analysis. Compared with CNT or CB, GR can be well wetted by NR and uniformly distributed in NR, so there was a strong interfacial interaction between them. The better interfacial interaction will promote the formation of a good network structure between the filler and the NR, and the network structure between the filler and NR further restricts the movement of rubber molecule chains. Therefore, the heat generation by the movement of the molecule chains is decreased, and the mechanical properties of the rubber composites are increased.


Introduction
Rubber has been widely used in industrial fields for the reversible deformation.However, the low mechanical properties such as tensile strength of rubber restrict its application.Therefore, it is necessary to add filler to enhance their mechanical performance.The most significant reinforcing filler in the rubber is carbon black (CB), which can remarkably improve the tensile strength, elastic modulus and compression resistance of rubber [1].The excellent mechanical properties of CB/rubber composites require a high addition of at least 30 phr CB into the nature rubber (NR) matrix, which makes it difficult to disperse CB during processing and causes a significant fatigue temperature rise in the rubber composite under cyclic loading [2], that fatigue temperature rise could decrease the mechanical properties of the rubber composites.Consequently, it is important to find new fillers to replace CB.
Carbon-based nanofillers, such as Graphene (GR), Carbon Nanotubes (CNTs), and others were utilized to create rubber composites with superior overall qualities, including improved mechanical properties, electrical and thermal conductivity [3,4].The main distinction between GR and CNTs is their dimension.CNTs are one dimensional nanomaterial with a high aspect ratio, in contrast to two dimensional GR.GR has a high specific surface area (~2600 m 2 /g) [4], which is significantly greater than CNTs (~100 m 2 /g) [5,6].However, a thorough knowledge of the different effects of CNTs and GR on the characteristics of rubber composites is still lacking.
The effects of CNTs or GR with different contents on the properties of CB/NR materials have attracted more and more attentions.By substituting one-dimensional CNT or two-dimensional flaky graphene oxide (GO) of the portion of CB, respectively, Guo et al. prepared CNTs-CB/NR and GO-CB/NR composites.Their findings demonstrated that the segregation effect between various fillers increased the dispersion of the composite fillers in the NR matrix, effectively preventing crack expansion and enhancing the fatigue life of the composites [6].Zhang and his coworkers determined the effects of 1 phr GO or multi-walled carbon nanotubes (MWNT) on CB/NR composites by replacing CB.Compared with CB/NR composites, NR composites with MWNT or GO have higher tensile strength, mainly due to their higher crosslink density, better filler dispersion, larger filler-rubber interaction, and higher effective elasticity [7].However, the action mechanism by which GO or MWNT were more effectively combined with NR interfaces was not fully described.The impact of GO and CNTs on heat generation of CB/NR composites also had not been deeply discussed.
Herein, the GR/CB/NR composites and CNTs/CB/NR composites were prepared.The influence of the synergistic effects between GR or CNTs with CB on the mechanical properties and fatigue temperature rise of NR composites was investigated.The reasons for the increased mechanical properties and decreased heat generation of CNTs/CB/NR and GR/CB/NR composites were discussed by contact angle.The network structure between the filler and the rubber matrix was observed by rubber process analyzer, and the glass transition temperature (Tg) was measured by dynamic mechanical analysis.

Preparation
The roller gap of two-roll mill was set to about 2 mm.NR was added and plasticized for 8 minutes, then GR, ZnO, SA, RD, 4010NA, CZ, and S were added one at a time.Next, there were 5 times of left-and right-side cuts, followed by a triangle package that was implemented 7 times and a thinpassing for 8 times.The compounds were then kept for 24 hours and cured for 8 minutes in a press vulcanizer at 150 °C.The vulcanization recipe is found in Table 1.

Measurements
According to ISO37-2011, a tensile tester (SANS, CMT 6104) operating at a rate of 500 mm ꞏ min-1 at room temperature was used to measure the mechanical properties.A Goodrich RH-3000N characterized the temperature increase of the sample under fatigue loadings.Cylindrical rubber specimens with dimensions of 17.5 mm in diameter and 25 mm in height were repeatedly compressed.The environmental temperature of the test was set at 55 °C.Thermocouples were used to gauge the temperature at the bottom and center of specimen.The frequency and pre-stress were 30 Hz and 1.0 MPa, respectively.The stroke was 4.445 mm.The test processes ended after 25 minutes (45000 cycles) of testing with a constant load.Three samples were measured for each individual composite, and the average values were calculated.
Dynamic mechanical analysis (DMA) was carried out using the NETZSCH DMA 242 instrument in the tensile mode with 0.5% dynamic strain.From -70 °C to 120 °C, the samples were heated at a rate of 3 °C/min at frequencies of 5 Hz.
Static contact angles of raw rubber and fillers were calculated using KRÜSS DSA 100 drop form studies (KRÜSS GmbH, Germany).By dissolving rubber with toluene, rubber film was produced.A few millimeters of CB, GR and CNTs film were created by compression molding on a machine for compressing powder.The test liquids included glycerol and deionized water, each with a different surface tension.To ensure data accuracy and reproducibility, three contact angles were measured for each sample, and the average value was calculated.
The Payne effect of the composites were examined using RPA S5 (SARTEC Testing Instruments Ltd., Guangdong, China) at 60℃.For the rubber compounds, the strain ranged from 0.1% -200% at the test frequency of 1 Hz.The test of each specimen was done three times.
The field-emission scanning electron microscopy (SEM) was utilized to observe the cryogenically fractured surfaces of the composites.(SEM, Gemini 500, Germany).

Mechanical properties
The results demonstrated that the partial substitution of CB by GR or CNTs has enhanced the tensile strength and tear strength of rubber composites (Figure 1).The tensile strength of CNTs/CB/NR and GR/CB/NR compared to CB/NR were increased by 8.47% and 12.49%, respectively.In addition, the tear strength of CNTs/CB/NR and GR/CB/NR were about 49.94% and 39.81% higher than those of CB/NR, respectively.For CNTs, their elongated tubular structure enables effective entanglement of rubber macromolecular chains on the surface of the CNTs.Furthermore, the presence of phospholipids at the end of the NR allows for robust interactions with the CNTs surface through cation- strong interaction [8].The rigid planar structure of GR enables it to retain its specific surface area more effectively compared to CNTs.Unlike CNTs, which are susceptible to bending and deformation, GR maintains its flat structure, allowing for enhanced interfacial interaction with the NR [8,9].As a result, CNTs and GR exhibit excellent interfacial interaction with NR, as discussed in section 3.3.

Fatigue temperature rise and heat generation
Figure 2a shows the fatigue temperature rise at the center (TC) and bottom (TB) of the sample.It can be seen that the temperature rise at sample center of the composite were significantly higher than the temperature rise at the sample bottom, because the heat dissipates more slowly at the center of the material compared with the bottom, and the heat accumulates internally.No matter the bottom temperature rise or the center point temperature rise, the fatigue temperature rise of the composites with the CB replaced by GR is lower than that of CNTs with the same content.The heat generated by the energy dissipation in the component ultimately increases the temperature rise of rubber composites.
The heat generation of every circle during the fatigue loadings can be obtained from the energy dissipation equation (1) [10].Figure 2b shows the heat generation by all cycles during fatigue, from which similar results of the fatigue temperature rise can be clearly seen.The heat generated by the GR/CB/NR composites was much less than that of the CNTs/CB/NR composites.This may be due to the better bonding between the GR and NR and the less internal heat generation due to friction.Strong interfacial interactions will effectively reduce the heat generation of the composite [11], thus the heat generation is smaller.
Where, ω is loading frequency, E" is the modulus of loss, δ is the hysteresis angle, and ε0 and σ0 are the maximum strain and maximum stress, respectively.

Microstructure
As mentioned earlier, the interfacial bonding between the filler and rubber plays a significant role in the properties of composites.The dispersion of the filler is closely linked to the interfacial bonding.Figure 3 presents the scanning electron microscope image of the liquid nitrogen brittle section of the composite material.The CNTs and GR exhibited a more homogeneous dispersion at a content of 3.5 phr, but some agglomeration was observed with an increase in content to 5 phr.

Interface interactions
The surface energy of a filler can be determined by measuring the contact angles between filler and rubber, according to equation ( 2) and equation ( 3) [10,11].In model of Fowkes, the surface energy can be divided into a polar component ( ) and a dispersive component ( ).
WFF and WRF relate to the work required to adhere the fillers and the filler-rubber, respectively.When WRF/WFF ≥1, the attraction between the filler and NR is greater than or equal to that between the filler and filler [12,13].This demonstrates how easily the NR may moisten the filler.Table 2 demonstrates that the largest value of WRF/WFF was GR, which indicated that GR can be well wetted by NR to be uniformly distributed in NR compared to CNTs or CB. (2) where, and are the polar and dispersive components of the surface energies of the filler, respectively.and are the polar and dispersive components of the surface energies of the rubber, respectively.are the total surface energies.
The physico-chemical compatibility of the filler surface and the rubber polymer is a crucial factor influencing the dispersion of the filler particles in the rubber polymers, which can be measured using the free energy of immersion (∆Gi) of the filler in the polymers (equation ( 4)).If ∆Gi is negative, it is thermodynamically beneficial for the rubber polymer to moisten the particles.Thus, the ∆Gi can be used to forecast the thermodynamic contribution to the dispersibility [12].A filler will disperse considerably more readily in a rubber polymer if the wetting of the filler particles by the polymer is thermodynamically favored, which is shown by a substantially negative ∆Gi value.Table 2 demonstrates that GR has the lowest ∆Gi value, which suggests that it disperses more readily in rubber than CB and CNTs.
The spreading work Ws is related to the mobility of the NR chains at the filler interface, which can be seen as a reflection of the interfacial interaction between the fillers and NR (equation ( 5)).The higher the value of Ws, the more appealing the interaction between fillers and NR [14].The GR had the greatest Ws value indicated that there was a stronger connection between GR and NR.
All the results show that there is a strong interfacial interaction between GR and NR compared with CNTs or CB, so that it can be well dispersed in the NR matrix.It is the strong interfacial interaction between them that can effectively inhibit the movement of NR molecule chains and therefore reduce the heat generation due to the friction of molecule chain movement.

Payne effect
A strain scan test was run on the composite in order to examine the strength of the filler network inside the composites.Figure 4 depicts the correlation between the elastic modulus (G') and strain of rubber composites.The graphic shows that G' rapidly decline as strain increases.This is because when the strain gradually increases, the filler network was damaged, releasing some of the rubber chains which were held together by the filler [15].The Payne effect, which refers to a nonlinear trait in the storage modulus of materials that decreases with the amplitude of dynamic strain, is a term used to describe the amplitude dependency of the dynamic properties of composites.The Payne effect is represented as ∆G'=G'0.1%-G'10%[11].The larger the ∆G', the higher the network degree of the filler and the poorer the interfacial interaction between the filler and the rubber.On the contrary, the smaller the ∆G', the lower the network degree of the filler and the stronger the interfacial bond between the filler and rubber [16,17].The ∆G' of GR/CB/NR composite were smaller than that of CNTs/CB/NR composites, which indicated a higher degree of network between the filler and the rubber of GR/CB/NR composites.Because the high degree of network between the filler and NR can effectively inhibits the movement between them, the heat generation by the friction of movement can be reduced.
Then GR/CB/NR composites had lower heat generation than CNTs/CB/NR composites.

Glass transition temperature of composites
Figure 5a is the loss factor of the composite.The maximum value of tan is related to the Tg.The Tg of each content composite was calculated using the temperature that corresponds to the maximum value of tan in Figure 5a.As Tg increases, the molecule chain flexibility decreases, leading to an increase in the stiffness of the composite, and therefore the tensile properties are enhanced.Also, the friction due to molecule chain movement decreases, then the heat generation from friction decreases.The lower deformation and lower friction all decrease the ability of the system to generate heat.GR/CB/NR composites with content 5.0-0.0 exhibit the highest Tg as illustrated in Figure 5b.This indicated that the 5.0-0.0GR/CB/NR composite was more difficult to move the molecule chains compared to other composites, and therefore the heat generation attributed from the movement of molecule chains was also reduced.

Conclusion
In conclusion, the effects of CNTs and GR with different ratios on the properties of CB/NR materials were investigated.Compared with CB, GR and CNTs combine more strongly with NR.Thus, the partial substitution of CB by GR and CNT has enhanced the tensile strength and tear strength of the rubber composites.Compared with CNT or CB, GR can be well wetted by NR and uniformly distributed in NR.The GR form higher degree of network with NR, correspondingly, the Tg is higher, which can reduce the heat generation due to the limited movement of NR chains.

Figure 1 .
Figure 1.Tensile strength and tear strength of the composite.

Figure 2 .
Figure 2. (a) Diagram of temperature rise of bottom and center points of rubber composites.(b) Heat generation of rubber composites.

Figure 4 .
Figure 4. Elastic modulus (G') as a function of strain of rubber composites.

Figure 5 .
Figure 5. (a) Loss factor and (b) Tg of the rubber composites.

Table 2 .
Surface properties of different fillers.