Mechanical and rheological properties of polystyrene-block-polybutadiene-block-polystyrene copolymer reinforced with carbon nanotubes: effect of processing conditions

2.1 Materials

The materials used in this work were:

• Styrene-b-butadiene-b-styrene copolymer (SBS), with styrene content at 30 wt%, purchased by Sigma-Aldrich (Saint-Louis, MO, USA). The main properties of used SBS are listed in Table 1.

• Bare multiwalled CNTs, purchased by Cheap Tubes (Grafton, VT, USA); main properties: OD=120÷180 nm, ID=10÷20 nm, L=10÷20 μm, purity >95 wt%, ash <1.5 wt%, specific surface area SSA >40 m²/g, and electrical conductivity EC>10⁻² S/cm.

• Multiwalled CNTs containing ~1 wt% of covalently linked –COOH groups (COOH–CNTs), purchased by Cheap Tubes, USA; main properties: outer diameter OD=120÷180 nm, inner diameter ID=10÷20 nm, length L=10÷20 μm, purity >95 wt%, ash <1.5 wt%, specific surface area SSA >60 m²/g, and electrical conductivity EC >10⁻² S/cm.

2.2 Processing

The preparation of SBS/CNTs and SBS/COOH–CNTs nanocomposites was carried out using a Brabender mixer (Duisburg, Germany) at three different temperatures, particularly at 120°C, 150°C and 180°C, and 50 rpm for 5 min. In order to evaluate the effect of the mixing speed, the nanocomposite SBS/COOH–CNTs obtained at 180°C was prepared at two different processing speeds, namely 50 and 100 rpm. The nanotubes were added at 3 wt% into the melted matrix after 2 min of processing. The chosen CNTs content ensures significant improvement of mechanical properties, avoiding re-aggregation phenomena of CNTs. The pristine SBS copolymer was subjected to the same processing.

The SBS and SBS-based nanocomposite specimens used for different characterizations were prepared by a compression molding step, using a Carver Press (Wabash, IN, USA) at the same processing temperature, for 5 min, under a pressure of about 30 MPa.

2.3 Characterizations

Rheological tests were performed using a strain-controlled rheometer (mod. ARES G2 by TA Instrument, New Castle, DE, USA) in parallel plate geometry (plate diameter 25 mm). The complex viscosity (η*) and storage (G′) and loss (G″) moduli were measured performing frequency scans from ω=10⁻² to 10² rad/s at same processing temperatures. The strain amplitude was γ=2%, which preliminary strain sweep experiments proved to be low enough to be in the linear viscoelastic regime.

Mechanical tests of the samples were carried out using a universal Instron machine (model 3365, High Wycombe, UK), according to ASTM D882 (crosshead speed of 100 mm/min). The average values for elongation at break, EB, and for elastic modulus, E, were calculated and the standard deviation is reported in the figures.

Dynamic mechanical thermal analysis (DMTA) was performed using a Rheometrics DMTA V instrument (Piscataway, NJ, USA) in round shear sandwich mode. The tests have been carried out using a temperature sweep method, between 30°C and 140°C at a heating rate of 5°C/min. The frequency was set to 1 Hz and the maximum strain amplitude was 2%. The storage modulus (G′) and the damping factor (tan δ) as a function of the temperature have been recorded.

Differential scanning calorimetry (DSC) was performed using a DSC 60 Shimadzu calorimeter (Kioto, Japan). All experiments were run with samples of about 10 mg in 40 μl sealed aluminum pans. The calorimetric scan (−110°C–105°C) was performed for each sample at a heating rate of 10°C/min.

The dispersion of CNTs in SBS matrix has been evaluated through scanning electron microscopy (SEM) observations, performed on nitrogen-fractured radial surfaces of the investigated samples with a Philips ESEM XL30 microscope (Eindhoven, The Netherlands) and transmission electron microscopy (TEM) with a Jeol JEM-2100 (Akishima, Japan) at 200 kV, on ultrathin films with thickness of about 100 nm, prepared via cutting from specimens block with a Leica Ultramicrotome EMUC6.

3.1 Effect of the processing temperature

Rheological analyses can provide useful information about the state of dispersion of nanoparticles within host polymer matrix [30]; furthermore, the assessment of the rheological behavior allows to evaluate the effect of the processing parameters on the internal structure of the nanocomposites. In Figure 1, the complex viscosity (η*) and the storage (G′) and loss (G″) moduli as a function of frequency for neat SBS and SBS-based nanocomposites formulated at different processing temperatures are reported. It can be observed that, as the processing temperature increases, the difference between the rheological behavior of the nanocomposites and that of the neat matrix becomes more pronounced. Indeed, for the systems formulated at 120°C and 150°C the complex viscosity and moduli values of the nanocomposites are quite similar to those of neat SBS copolymer, and no variation of the rheological behavior attributable to the CNTs presence can be detected. These features seem to indicate a poor dispersion of CNTs in these systems [31]. Differently, the nanocomposites formulated at 180°C show higher complex viscosity and moduli values than those of neat SBS in the whole investigated frequency range, and this behavior is more pronounced for COOH–CNTs containing nanocomposite. The last can be understood considering that the functionalization of CNTs with carboxylic groups enhances the affinity between matrix and nanofillers, improving the dispersion of CNTs within SBS copolymer. Additionally, the disappearance of the Newtonian plateau for nanocomposites containing both unmodified and –COOH functionalized CNTs can be noticed. The obtained results suggest that the processing of the nanocomposites at 180°C leads to the obtainment of a more uniform and homogeneous distribution of nanofillers within host matrix with respect to the nanocomposites obtained at lower processing temperatures. Additionally, the differences in the rheological behavior between the systems formulated at different temperatures could be explained considering some modification of the internal structure of the neat SBS copolymer. Indeed, it is well known that in block copolymers, as a function of the processing temperature, modifications of the internal microstructure can occur and different morphologies and orientation of the domains can be obtained [32], [33].

Figure 1:

Complex viscosity and storage (full symbols) and loss (empty symbols) moduli as a function of frequency for neat SBS and SBS-based nanocomposites at different processing temperatures.

In Figure 2, the main mechanical properties, e.g. elastic modulus (E), tensile strength (TS), and elongation at break (EB) for neat matrix and nanocomposites formulated at different temperatures are reported. Concerning the mechanical behavior of neat SBS copolymer, a progressive increase of E and a decrease of EB can be observed with an increase of processing temperature; differently, the TS values seem to be almost unaffected by the processing temperature. As discussed earlier, this behavior could be attributed to the variation of the internal microstructure of SBS copolymer as a function of temperature [32], [33].

Figure 2:

Main mechanical properties for neat SBS and SBS-based nanocomposites: (A) elastic modulus, (B) tensile strength, and (C) elongation at break as a function of processing temperature.

Let us now consider the effect of the nanofillers addition. Irrespective of the processing temperature, the presence of CNTs causes the increase of the elastic modulus and the decrease of the elongation at break, indicating that the nanocomposite rigidity is improved as a result of the nanofillers addition while the nanocomposites become somewhat brittle compared to neat SBS. The presence of –COOH functionalized CNTs leads to the obtainment of superior tensile properties with respect to unmodified CNTs-containing nanocomposite. The last can be understand considering that the presence of functional groups onto the CNTs surface enhances the interfacial adhesion between macromolecular chains and dispersed nanoparticles, promoting in this way the stress transfer across the interface.

As already noticed in the analysis of the rheological behavior, the differences between nanocomposites and neat matrix become more pronounced as the processing temperature increases. For instance, the elastic modulus of SBS/COOH–CNTs nanocomposites formulated at 120°C and 180°C increases by 9.5% and 37%, respectively, compared to the neat matrix obtained at the same temperatures. The obtained results suggest that the increase of the processing temperature has a beneficial effect on the dispersion of nanofillers, as the improved mechanical properties showed by nanocomposites formulated at 180°C are consistent with the achievement of a more finely and uniform dispersion of CNTs throughout the copolymer matrix.

To deeply investigate the mechanical behavior of SBS-based nanocomposites, dynamical-mechanical tests have been carried out in the viscoelastic region. In Figure 3, the trends of dynamic storage modulus (G′) and damping factor (tan δ) as a function of temperature are reported, for neat SBS and CNTs-containing nanocomposites obtained at three different processing temperatures. All investigated systems show a sudden fall in the storage modulus at a temperature around 100°C corresponding to a relaxation process involving the polystyrene domains [34]. Once again, the CNTs addition results in significant variation of the SBS mechanical behavior solely for the nanocomposites formulated at 180°C, for which an increase of the G′ value in the whole investigated temperature range corresponding to an increase of the nanocomposite rigidity can be observed. This increase is exacerbated for the nanocomposite containing COOH–CNTs, due to the presence of functional groups that improve the compatibility between nanoparticles and SBS chains, enhancing the interfacial adhesion and the nanoparticles distribution. Differently, the nanocomposites obtained at lower temperatures show a G′ trend as a function of temperature similar to that of neat matrix.

Figure 3:

Dynamic storage modulus and damping factor for neat SBS and SBS-based nanocomposites obtained at different processing temperatures.

In correspondence of the relaxation process recognized in G′ trend, a well-distinguished maximum in the tan δ curve can be observed, related to the glass transition of polystyrene portion [34]. Regardless of processing temperature, the addition of CNTs and, even more, of COOH–CNTs causes a shift of tan δ peak toward higher temperatures, indicating the loss of macromolecular mobility of SBS chains. The presence of CNTs, indeed, hinders the local motion of copolymer chains [35] and induces a lowering of the viscous energy associated with the macromolecular motions, leading to an increase of the SBS glass transition temperature.

To better evaluate the effect of the processing temperature and of the CNTs adding on the glass transition temperature (Tg) of SBS copolymer, DSC analyses have been carried out and the obtained results have been listed in Table 2 (Mixing speed 50 rpm). Furthermore, in Figure 4 the thermograms collected during the heating scan for neat SBS and SBS-based nanocomposites processed at 180°C are reported. Neat SBS, irrespective of the processing temperature, shows two different endothermic peaks: one at about −95°C, which corresponds to the Tg of polybutadiene domains, and one at about 74°C, related to the Tg of the polystyrene phase. The addition of bare and functionalized CNTs leads to an increase of both Tg values, because of the decreased mobility of SBS macromolecules induced by the CNTs presence, as already inferred from the analysis of nanocomposites thermo-mechanical properties.

Figure 4:

DSC thermograms for neat SBS and SBS-based nanocomposites processed at 180°C.

The SEM micrographs of fracture surface of SBS-based nanocomposites prepared at three different processing temperatures are shown in Figure 5. It is evident that in the nanocomposites formulated at 120°C and 150°C a poor dispersion of CNTs has been achieved. Indeed, the presence of nanofillers agglomerates can be noticed in both unmodified and functionalized CNTs-containing systems, although in the SBS/COOH–CNTs nanocomposites a more regular morphology is noticeable. It seems that, due to the improper processing temperature, the nanofillers are selectively segregated in the polybutadiene and/or polystyrene domains of the SBS matrix. Otherwise, the nanocomposites formulated at 180°C show a homogeneous and uniform dispersion of the nanofillers within the host matrix; moreover, the lack of CNTs pulled-out from the matrix indicates the strong level of interfacial adhesion reached in the SBS-based nanocomposites obtained at 180°C. To better evaluate the morphology of the SBS-based nanocomposites processed at 180°C, TEM analysis has been performed, see micrographs reported in Figure 6. As already inferred from the analysis of SEM results, both bare and COOH-functionalized CNTs s are uniformly distributed within SBS host matrix, indicating that the process at higher temperature is effective in the enhancement of the nanofillers dispersion.

Figure 5:

SEM micrographs of all investigated SBS-based nanocomposites.

Figure 6:

TEM micrographs of nanocomposites processed at 180°C.

3.2 Effect of the mixing speed

In order to evaluate the effect of the mixing speed on the final properties of SBS-based nanocomposites, the processing of neat SBS and SBS/COOH–CNTs nanocomposites obtained at 180°C has been carried out at 50 and 100 rpm. In Figure 7, the complex viscosity curves and the trends of G′ and G″ moduli as a function of frequency are reported. It can be clearly observed that the mixing speed has a negligible effect on the rheological function of both neat SBS and COOH–CNTs-containing nanocomposite. Indeed, either the complex viscosity curves or the moduli trends of SBS-based systems obtained at 50 and 100 rpm are almost coincident, highlighting that the mixing speed is not the determinant in the modification of the rheological behavior of the investigated systems.

Figure 7:

Complex viscosity and storage (full symbols) and loss (empty symbols) moduli as a function of frequency for neat SBS and SBS/COOH–CNTs nanocomposite at different mixing speeds.

Similarly, the mechanical behavior of neat SBS and SBS/COOH–CNTs nanocomposites is slightly affected by the mixing speed. In Figure 8A–C, the values of E, TS, and EB for the two systems processed at 50 and 100 rpm are reported. It can be noticed that the values of the mechanical properties for the neat matrix obtained at 100 rpm are scarcely lower than those showed by SBS processed at low mixing speed, suggesting that the neat matrix can experience thermo-mechanical degradation at such high mixing speed. Concerning the mechanical properties of COOH–CNTs-containing nanocomposite, the increase of the mixing speed causes the enhancement of E and TS and a decrease of EB, highlighting that the processing at high mixing speed promotes a more uniform dispersion of CNTs within host matrix.

Figure 8:

(A) Elastic modulus, (B) tensile strength, (C) elongation at break, (D) dynamic storage modulus for neat SBS and SBS/COOH–CNTs nanocomposite at different mixing speeds, and (E) SEM micrographs of SBS/COOH–CNTs nanocomposite obtained at high mixing speed.

The same conclusions can be drawn from the analysis of the thermo-mechanical behavior of neat SBS and COOH–CNTs-based nanocomposite as reported in Figure 8D. Looking at the trends of the dynamic elastic modulus as a function of the temperature for neat SBS and CNTs–COOH-containing nanocomposite, a decrease of G′ for SBS and an increase of the modulus values for the nanocomposite as a function of the mixing speed can be observed.

As far as the thermal properties of neat SBS and SBS/COOH–CNTs nanocomposite processed at 50 and 100 rpm are concerned, looking at the Tg values reported in Table 2 (Processing temperature 180°C, mixing speed 100 rpm), it is clearly observable that the variation of mixing speed has a negligible effect on the glass transition temperatures of both neat SBS copolymer and COOH–CNTs-containing nanocomposites.

In Figure 8E, the SEM micrograph of the SBS/COOH-CNTs nanocomposite processed at 100 rpm is reported. A homogeneous and uniform dispersion of functionalized CNTs within host matrix is noticeable, although no remarkable differences with respect to the morphology of the nanocomposite obtained at 50 rpm can be observed.