3.1. Scanning electron microscope (SEM)
Figure 3 shows the SEM morphology of the cryo-fractured PVDF-SBR blend matrix and its nanocomposites for GOn/Fe3O4 at (2.5/2.5). The neat PVDF-SBR (65/35) blend shows a plain type morphology (absence of nanofillers).
It can be easily seen that the morphology of neat PVDF-SBR blend polymer is a sea-island morphology with a large amount of droplet phenomenon and droplet size close to 40 µm which confirms that both polymers are immiscible due to the rough appearance of the rubber component of the SBR [2, 16]. On the other hand, Fig. 3b, shows that the addition of nanofillers ( GOn-Fe3O4) at 2.5 wt.% may decrease the size of droplets to the average diameter of 18µm leading to a formation of finer morphology with the presence of lower amount of pullout phenomenon as compared of neat PVDF-SBR material that may be ascribed to compatibilizing effect of nanofillers via the physical interaction. The morphology of hybrid nanocomposites starts to transform in a co-continuous structure with the presence of highly dispersed a small droplet with a size of 18 µm and the appearance of some dispersed nanosheet ( GnO) and nanoparticles ( Fe3O4) of PVDF-SBR with a size around of 7 − 6 µm. This structure is due to the coalescence suppression of phase-separated domains trough the formation of physical interaction between the polymer components via nanofillers [30]. For the Fig. 3 (b’), the morphology of the hybrid nanocomposite PVDF-SBR/GOn-Fe3O4 at (2.5/2.5) sample shows that good dispersion/distribution of the nanofillers was obtained as no traces of agglomeration can be seen. The presence of agglomeration would indicate poor affinity and lower distribution homogeneity [31, 32]. In our case, the GOn and Fe3O4 nanoparticles are homogeneously distributed due to the functional group interaction between both nanofillers. The oxygen content in the functional groups on the GOn surface can interact with the Fe3O4 nanoparticles via hydrogen bonding as well as van der Waals interactions, resulting in good dispersion of the nano-fillers inside the PVDF. Moreover, it is clearly shown on Fig. 3(a’,b’) that good interfacial adhesion was produced after the nanofillers incorporation since no deboning can be seen. Finally, the nanofillers are well dispersed inside the matrix and it can be concluded that the masterbatch approach and the processing conditions used were optimum leading to good dispersion/distribution and uniform nanocomposite properties[33].
3.2. Fourier transform infrared spectroscopy (FTIR)
Fourier transform infrared (FTIR) spectroscopy is an effective method to study possible crystalline phase transformation in PVDF. Figure 4 shows the FTIR spectra of the neat PVDF-SBR blend and the PVDF-SBR nanocomposites at different GOn and Fe3O4 content. The FTIR spectra of the neat matrix is composed of the two main characteristic bands of both PVDF and SBR. The characteristic bands of SBR are located at 3040 − 3020 cm− 1 corresponding to C–H stretching, the band at 1479 cm− 1 corresponds to CH2 in-plane deformation, and 1376 cm− 1 band is characteristic of –CH2 wagging motion; while the PVDF has its main characteristic bands at 975, 720, 672, 659 and 469 cm− 1 which are specific to the presence of the pure α-phase. The bands at 975 cm− 1 corresponds to CH2 rocking, the band at 720 cm− 1 is characteristic of bending and skeletal bending, the bands at 672 − 659 correspond to CF2 groups and rocking vibration, and the band at 469 cm− 1 is associated to CF2 bending. From this analysis, it is clear that the PVDF blended with the SBR is only only showing the α-phase, similar to the pure PVDF.
It is clear from the FTIR spectra of the nanocomposites that new bands are presents compared to those of the PVDF α-phase, depending on the GOn, Fe3O4 or GOn-Fe3O4 content (wt.%). They are localized at 1218 − 1178, 966, 907, 869, 600, 568, and 437 cm− 1 which are assigned to the β-polymorph structure of PVDF. This confirms that nanoparticles addition led to some PVDF crystal transformation from the α-phase to the β-phase as reported in the literature [27]. The increase of the β-phase content in the nanocomposite can be related to the presence of GOn and Fe3O4 nanoparticles providing additional nucleation sites (heterogeneous nucleation) for the crystallization of the β-phase. The formation of the β-polymorph in the PVDF-SBR nanocomposites is also attributed to the strong and specific interaction between the carbonyl groups in graphene oxide and the fluoride groups in PVDF (-C-O) found in graphene oxide and the CF2 segments of PVDF leading to a transformation from the α-phase’s trans-gauche-trans-gauche conformation into the trans-trans conformation characteristic of the β-phase. These results confirm the presence of a mixture of α- and β-polymorph in the PVDF-SBR nanocomposites studied. Among the observed polymorphs, the polar β phase of PVDF is the most desirable form because of its high permittivity making it a good candidate for piezoelectric material based on polymer matrices[27, 34].
3.3. Thermal properties:
The thermal resistance of the PVDF-SBR blend and its nanocomposites was evaluated using TGA. Figure 5 presents the weight curves a function of temperature at a heating rate of 10°C/min in air. It can be seen that all the curves have a similar shape, but the thermal stability increases with the presence of the nanoparticles. This is especially the case when compared with the neat PVDF matrix in the temperature range of 300–800°C. The nanocomposites filled with Fe3O4 alone (5 wt.%) show the highest thermal stability compared to the nanocomposites filled with GOn alone at the same content. This can be related to the higher intrinsic thermal stability of iron oxide compared to the graphene oxide nanoplatelets [35].
The temperature associated to the maximum degradation rate (Tmax) was obtained by the position of peak in the derivative thermogravimetric curve (DTG). Based on the DTG curve of the neat matrix (PVDF-SBR), it is clear that two maxima are present, which are associated to both polymers; the first one corresponds to SBR (437°C) and the second one is PVFD (464°C). It is clear in Fig. 5 that the Tmax of the nanocomposites is higher than that of the neat PVDF-SBR matrix. The Tmax of PVDF-SBR matrix is 464°C and increases to 470 oC and 477°C for the 5 wt % GOn and Fe3O4 nanocomposites, respectively. On the other hand, Tmax of the hybrid system (GOn:Fe3O4 at 2.5:2.5) shows an intermediate value (472°C). From these results, it clear that the incorporation of GOn and Fe3O4 nanoparticles in the PVDF-SBR blend matrix induced better thermal stability and thus the starting decomposition temperature is shifted to higher temperatures. During thermal degradation, GOn and Fe3O4 nanofillers, due to their inorganic nature, with the homogeneous distribution and distribution of nanofillers prevent the evolution of volatile decomposition products acting as a barrier inhibiting the propagation of heat from the environment in the polymer matrix [36, 37]. On the other hand, there is a close relationship between the crystallin phase of PVDF and the obtained results regarding the increased thermal stability. The addition of GOn or Fe3O4 (alone or together) leads to additional nucleation sites for the crystallization of the β-phase of PVDF leading to PVDF crystal transformation from the α-phase to the β-phase. In general, the crystaline phase has a higher thermal stability compared to the amorphous phase due to the ordered structure leading to a highest energy dissipation [27]. Finally, GOn and Fe3O4 can act as nucleating agent, as well as thermal stabilizers.
3.4. Tensile testing
The large aspect ratio of the GOn and Fe3O4 nanoparticles and their respectively high Young’s modulus should have a significant reinforcement effect on the mechanical properties of the PVDF-SBR blend [3, 22, 23]. The homogenous dispersion/distribution of the nanofillers into the matrix, along with favorable interfacial interactions between the nanofillers and the polymer blend, as well as the high level of exfoliation of GOn are the key points to achieve polymer nanocomposites with highly enhanced final properties[40, 41].
The tensile properties of neat PVDF-SBR blend and its nanocomposite were investigated by uniaxial tensile testing. Figures (6) presents the Young’s modulus, tensile strength, and strain at yield as a function of GOn and Fe3O4 content. Firstly, the neat PVDF, neat SBR and PVDF-SBR blend tensile properties are reported in Table 2. From these results, it is clear that SBR has better elasticity (higher strain at yield) compared to PVDF which is more rigid (higher modulus and strength). So the properties of the blend are in between. In our case, one of the main goal is to produce rigid nanocomposites while maintaining some ductility. So Fig. 6a shows that the addition of GOn and Fe3O4 nanoparticles gradually changes the tensile properties of the PVDF-SBR blend for the range of Fe3O4 loading (1.25-5 wt.%) investigated. But the nanocomposite containing only 5 wt.% Fe3O4 showed better mechanical properties with respect to the nanocomposite containing 5 wt.% GOn and the neat matrix. The Young’s modulus of these nanocomposites (1.25, 2.5 and 5 wt.% Fe3O4) was increased by around 85% compared to the neat PVDF-SBR matrix and the nanocomposite with 5 wt.% GOn. Higher Young’s modulus with Fe3O4 is due to the intrinsic properties of this nanoparticles compared to the nanosheet in terms particle shape. The sheet shape of GOn prevents their dispersion and distribution into the polymeric matrix at higher loading. So the GOn start to aggregate for content above 2.5 wt.%, while for Fe3O4 with their particulate shape (more spherical), there is a good dispersion and distribution into the matrix even up to 5 wt.% [42, 43].
When particle hybridization was performed by combining Fe3O4 and GOn, it can be seen that all the nanocomposites exhibit superior Young’s modulus than for the nanocomposites based on GOn or Fe3O4 alone. This confirm that a synergistic effect occurs between both particles leading to better reinforcement of the PVDF-SBR matrix. The results also show that by decreasing the GOn content and increasing the Fe3O4 nanoparticles (fixed total content), the Young’s modulus gradually increase to reach a maximum value (2052 MPa) at a GOn:Fe3O4 of 2.5:2.5 wt.% before decreasing. So this hybrid nanocomposites presents the optimum results due to the synergetic effect induced by the combination of GOn and Fe3O4 [23],[44]. Such synergistic interactions result in the formation of an homogeneous interconnected network structure, which explains the large increase in the matrix Young’s modulus. On the other hand, the phenomenon of nanosheets restacking at GOn:Fe3O4 of 2.5:2.5 wt.% does not occur and the nanosheets are still well dispersed and distributed within the matrix. Actually, the GOn surface becomes decorated with Fe3O4, which inhibited the sheet-to-sheet aggregations of GOn as evidenced by SEM observations in Fig. 3; i.e. no aggregation of sheets or nanoparticles was observed, ascribed to the good processing conditions used leading to high distribution and dispersion level of GOn and Fe3O4 combined with the strong hydrogen bonding between the matrix and the surface of GOn and Fe3O4 [44]. For the nanocomposites containing only one type of particle (GOn or Fe3O4) at 5wt.%, the relatively lower Young’s modulus can be associated to the possible agglomeration of some nanosheets. These agglomerates may prevent an efficient interfacial load transfer and changes in the fillers aspect ratio [23, 45].
Figure 6a also shows the tensile strength of the nanocomposites as a function of nanoparticles content. It is clear from Fig. 6a that a slight tensile strength decrease (5%) is observed for all the nanocomposites compared to the neat matrix. Generally in nanocomposites, an increase of the nanofillers content combined with a weak interfacial adhesion between the nanofillers and the matrix leads to poor interfacial quality and more decohesion zone are created in the nanocomposites under stress. This decohesion acts as stress concentration points accelerating sample failure [46]. On the other hand, it is observed that the tensile strength of all the nanocomposites are similar close to 44.7 MPa (± 0.2 MPa). This result may confirm the good interfacial adhesion generating better stress transfer under tensile load; i.e. better compatibility and affinity between GOn and Fe3O4 with the matrix. This is especially the case for the hybrid nanocomposites as the tensile strength is almost constant independent of the GOn:Fe3O4 ratio. The interfacial adhesion between the GOn:Fe3O4 hybrid system and the polymer matrix was very good as confirmed by SEM images (Fig. 3). As a results, the GOn:Fe3O4 hybrid nanofiller network acts as an efficient reinforcement in the parent system, providing superior tensile properties to the nanocomposites. In this case, effective stress transfer occurred from the polymer chains to the dispersed hybrid nanofiller, providing additional strength to the nanocomposites attributed to the synergistic effect generated from the combination of GOn and Fe3O4. Therefore, the successful addition of a relatively high nanofiller content into the polymer matrix strongly depends on the good compatibility and the strong interfacial interaction between all phases present.
Change in the ductile behavior of the nanocomposites related to GOn and Fe3O4 content and their hybridization was also investigated according to the strain at yield results. Figure 6b shows that the strain at yield of the nanocomposites gradually decreased with nanofillers loading. The strain at yield decreased to 20% for all the nanocomposites compared to the blend matrix. However, an almost contant value of the strain at yield values around 0.41 mm/mm (± 0.01 MPa) is observed for all the hybrid nanocomposites. Lower strain at yield for the nanocomposites was expected because of the intrinsic rigid character of both nanoparticles which have low deformation and act as stress concentrators accelerating crack initiation. The nanofillers used, even at lower content, also acted as nucleating agents reducing the amount of plastic energy (elasticity) absorbed by the nanocomposites under stress leading to lower ductility [47]. Another reason for the low ductility of the nanocomposites is due to the large aspect ratio and the strong interactions between the GOn and Fe3O4 with the matrix, which increased the crystallinity, both imposing restriction on the polymer chains mobility [48]. However, the almost constant strain at yield of all the nanocomposites, even with the increased of rigidity, is mainly due to the effect of the elastomeric nature of the SBR which confers the ductile behavior of the nanocomposites by preventing crack propagation along the interfacial area and facilitates mechanical energy dissipation. Furthermore, SBR addition led to maintain the strain at yield as it increases the interfacial adhesion improving the stress transfer to the nanofillers and promoting an efficient distribution of the applied stresses.
Such improvement in mechanical properties for the hybrid nanocomposites at a GOn:Fe3O4 ratio of 2.5:2.5 wt.% can be related to the good dispersion of GOn and Fe3O4, which is related to the high compatibility between all the phases, as well as the presence of strong hydrogen-bonding interactions between the polymer matrix chains and the nanofillers generating a strong synergy between GOn, Fe3O4, and PVDF-SBR. All these effects are required to improve the interfacial stress transfer from the polymer matrix to the individual nanofillers, thus increasing the nanocomposite mechanical properties.
Table 2
Tensile properties of the used polymers.
Polymer
|
Young’s modulus (MPa)
|
Tensile strength (MPa)
|
Strain at yield (mm/mm)
|
PVDF
|
1500
|
50
|
0.15
|
SBR
|
40
|
25
|
2.5
|
PVDF/SBR (65/35)
|
1098
|
46.9
|
0.5
|
3.5. Rheological properties:
The dynamic rheological properties of the nanocomposites were measured to provide further information on the internal structure and processability of these materials. Moreover, the information on the nanofillers dispersion state (percolated network structure), the effect of rigid particle addition on the motion of polymer chains, and interaction between the components can also be extracted.
Figure 7 compares the rheological properties of the nanocomposites in the melt state. It can be seen that the nanofiller content and frequency play an important role in the materials response to dynamic stress. Firstly, the addition of GOn or Fe3O4 into the polymer matrix disturbs the mobility of the polymer chains in the melt, thus increasing the storage modulus and loss modulus of the nanocomposites compared to the neat matrix. This also confirms the reinforcement effect of the nanofillers. The presence of rigid nanoparticles limits the polymer chain mobility and changes their molecular dynamics [49, 50]. Thus, large-scale polymer chain relaxation in the nanocomposites was effectively restrained by the presence of rigid inclusions. On the other hand, it is clear that the nanocomposites containing Fe3O4 alone (5 wt.%) produced significantly higher storage modulus than their GOn counterparts (5 wt.%). This behavior indicates that the nanofiller type has a direct effect of this property, as for the tensile properties (Fig. 5). This can again be related to the particle shape and their interactions with the matrix.
Figure 7 also shows that all the hybrid nanocomposites have intermediate values of storage modulus and loss modulus, between that of GOn and Fe3O4 nanocomposites. The synergetic effect induced by the combination of both particles is more effective in restraining the polymer relaxation than using GOn or Fe3O4 alone. This is due to the higher surface area and higher aspect ratio when two nanofillers are combined (better organization). Together with the high surface area of Fe3O4 and the nanoscale flat surface of GOn, the hybrid system led to stronger interfacial interactions with the matrix, good distribution/dispersion, and substantially higher effect on the polymer chain motion.
Figure 7 clearly show that the rheological properties of the nanocomposites have a predominantly elastic behavior at high frequencies (G’ > G’’), while a more viscous behavior is observed at low frequencies (G’ < G’’). This transition is associated to the polymer chains relaxation time [51]. The frequency also plays an important role on the nanocomposites' rheological response which is related to the viscoelastic behavior of these materials; i.e. a more solid-like behavior at higher frequencies as there is not enough time for the chain entanglements to occur and follow the deformation imposed, so a small amount of relaxation results in a higher value of the storage modulus (G′). On the other hand, the low-frequency region is related to reptation relaxation which is the motion of the whole chain [52].
One of the most important properties of thermoplastics materials is their viscosity which is a key property for successful manufacturing. The viscosity allows to predict flow of material inside processing equipment like extrusion and injection machines. It also allows to optimize the processing parameters (temperature, flow rate, screw speed, etc.) to produce homogeneous and stable parts [53]. Generally, viscosity is a function of the number of effective chains participating in the formation of a network structure. Figure 7c shows the complex viscosity as a function Gon and Fe3O4 content, and frequency. The results show an increase in complex viscosity with the addition of GOn at 5 wt.% to reach a maximum for the nanocomposites containing Fe3O4 nanofillers (5 wt.%). However, all the hybrid nanocomposites show intermediate values. The increased complex viscosity at low frequency of all the nanocomposites compared to the neat matrix can be attributed to the nanoscale dispersion/distribution of the nanofillers and to the presence of a higher number of interacting chains resulting from strong interfacial adhesion [54].
It can also be seen in Fig. 7c that the complex viscosity decreases with increasing frequency due to the strong shear-thinning behavior of polymers in the a melt state [55, 56]. A significant increase in the complex viscosity (η*) of the nanocomposites with increasing GOn and/or Fe3O4 content can be seen. The effect of the nanofillers content is mostly seen at low frequency, and the relative effect decreases with increasing frequency because of shear thinning behavior. The entanglement of polymer chains hinders the shear flow at a lower frequency, therefore higher viscosity is observed. But lower viscosity makes easier the processing and blending of nanocomposites at higher nanofiller content.
3.6. Electrical properties:
One of the key challenges to fabricate thermoplastic nanocomposites with high electrical and mechanical properties is related to the good processing conditions leading to the production of an interconnected network of nanofillers. Electrical percolation in conducting polymer nanocomposites containing nanofillers is strongly dependent on the properties of the nanofiller, namely their aspect ratio (L/D), electrical properties, and dispersion, as well as the nanocomposite microstructure associated to the dispersion and orientation of the nanofiller within the polymer matrix [57]. The addition of GOn and/or Fe3O4 has an effect on the electrical conductivity of the polymer matrix because of the intrinsic electrical properties of graphene and iron oxide which can provide percolated pathways for electron transfer making the nanocomposites electrically conductive even at low content [58, 59].
The effect of nanofillers on the electrical conductivity of the PVDF-SBR matrix is shown in Fig. 8. From the results obtained, it is clear that the electrical conductivity was influenced by the addition of 5 wt.% GOn, 5 wt.% Fe3O4 and 5 wt.% GOn:Fe3O4 hybrid nanofillers at various ratios. The electrical conductivity in Fig. 8 show an increases from 4.37x10− 13 S/m for the neat PVDF-SBR matrix to 1.71x10− 11 and 7 x10− 12 S/m for 5 wt.% GOn and 5 wt.% Fe3O4 nanocomposite, respectively. The nanocomposites with GOn alone had higher electrical conductivity than nanocomposites with Fe3O4 alone because of the higher aspect ratio of GOn. The percolation theory describes the connectivity of objects within a network structure and the effects of this connectivity on the macroscale properties of the system [60]. The nanofillers with nanoscopic dimensions and thereby large surface area to volume ratios tend toward agglomeration due to the strong interparticle attractive forces. This is particularly true for GOn that experience high van der Waals interlayers interactions. This intrinsic attraction, coupled with their high aspect ratios, can lead to substantial GOn aggregation. In addition to the nanoscale bundling or clustering of the nanofillers, the micro- and macro-scale dispersion/distribution is affected by both the internal inter-nanofillers and nanofillers interactions and by the external forces applied during the nanocomposite fabrication. The details of the nanofillers microstructure significantly affect the electrical properties of the final nanocomposite and give more advantage to the nanocomposites based on GOn.
The electrical conductivity of the nanocomposites containing hybrid nanofillers is higher than Fe3O4 nanocomposites at the same nanofiller content. However, the results show that the hybrid nanocomposites GOn:Fe3O4 -2.5:2.5 has an intermediate value (1.10x10− 11 S/m). As a result, the synergistic effect between GOn and Fe3O4 makes it possible for the partial replacement of the high aspect ratio and high-cost nanofillers (GOn) with low aspect ratio and lower cost Fe3O4.
3.7. Magnetic properties
Figure 9 presents the magnetic properties of the nanocomposites as measured by the system described in Fig. 2. The set-up determines the force (stress) necessary to detach the nanocomposites from a magnet. The measured magnetic stress presented in Fig. 9 shows that all the nanocomposites containing Fe3O4 alone at 5wt.% and GOn:Fe3O4 hybrid nanofillers exhibit superior magnetic properties than those observed for the nanocomposite containing GOn at 5wt.% alone. The magnetic stress increases linearly from 0 Pa for the GOn nanocomposites to reach a maximum of 870 Pa for the Fe3O4 nanocomposites at 5 wt.%. This increase with Fe3O4 content confirms the ferromagnetic behavior of the nanocomposites, which is attributed to the magnetic properties of the Fe3O4 nanoparticles and to the good dispersion/distribution of the magnetic nanofillers inside the polymer matrix confirmed by SEM (Fig. 3) and mechanical (Fig. 6) results [4, 61].
For the hybrid nanocomposites, the magnetic stress increased with increasing Fe3O4 ratio. At low nanofillers content, the nanocomposites are superparamagnetic due to the directions of weak magnetization axes which are randomly distributed [62, 63]. However, at high concentration, a interaction between the nanofillers occurs to form an infinite conductive network leading to ferromagnetic properties. The advantage of introducing magnetic nanoparticles in a material is twofold, as magnetic field gradients can be used to move the material, while alternating magnetic fields can be used to locally heat up the regions in the proximity of Fe3O4 .As the size of magnetic particles is reduced below a critical diameter (typically in the order of tens of nm), particles behave as superparamagnets[64] i.e., the atomic moments of the nanoparticle are aligned into a giant magnetic moment.
Once again for the magnetic properties, the hybrid system is an excellent way to produce nanocomposites with good magnetic properties, but good distribution/dispersion of the nanofillers in the matrix is required, which also leads to higher mechanical properties.