Effect of viscosity and processing parameters on the electrical conductivity of blends

8 July 2016
Kejing Yu, Aboutaleb Ameli, Yasamin Kazemi, Sai Wang, and Chul B. Park
An investigation into polypropylene/polystyrene blends shows thermodynamic factors to be more influential than kinetic ones in selectively localizing carbon nanotubes.

Carbon nanotubes (CNTs) possess a number of outstanding properties, in particular, fast charge transfer across interfaces. As a result, CNTs have been widely used as a new, ideal type of carbon-based electrically conductive filler for conductive polymer composites (CPCs).1, 2 CPCs themselves are used in packaging, insulation, sealing, and in materials for integrated circuits. However, CPCs have a high electrical percolation (a spike in conductivity related to filler concentration) threshold owing to their single-polymer phase.

It is thought that by introducing co-continuous immiscible polymer blends into CPCs, it may be possible to significantly decrease the electrical percolation threshold.3 This is the case if the electrically conductive filler percolates within one of the polymer phases, or at the interface.3 Until now, however, the mechanism and driving force of the selective localization of fillers in binary polymer blends has proved complicated.4 This is because of competition between the kinetic factors (melt viscosity ratio, mixing time, shear force, sequence of incorporation) and thermodynamic factors (interfacial and wetting coefficients).

Here, we describe the effect of kinetic factors on the morphology and electrical properties of polypropylene/polystyrene (PP/PS) blends containing CNTs (PP-CNTs/PS).5 We prepared these blends by melt mixing in a DSM twin-screw compounder. The blends differed in the melt flow rate of the PS (i.e., 10, 13, and 18g/10min) and we varied the kinetic factors for all the blends (see Table 1). We then cut the blends into pellets and prepared the samples for melt rheology and electrical conductivity testing by compression molding (2mm-thick plates, 5kN, 200°C). The melt viscosity ratios of the PP-CNTs/PS blends are given in Table 2.

Processing conditions used in the experiments.

TermsBlendsCompounding conditions
Melt viscosityPP-CNTs PS-10/13/18Simultaneously, 200°C, 10min, 75/100rpm
Sequence of incorporationPP-CNTs PS-18Simultaneously, 200°C, 10min, 75/100rpm
PP-CNTs or PS first 200°C, 10min, 75/100rpm
Mixing timePP-CNTs PS-18Simultaneously, 170°C, 2/5/10/15min, 200rpm
Simultaneously, 200°C, 2/5/10/15min, 200rpm

Melt viscosity (η) ratio of the PP-CNTs/PS blends.

FrequencyηPP−CNTsPS−10ηPP−CNTsPS−13ηPP−CNTsPS−18
75rpm2.642.291.96
100rpm2.412.091.83

We tested the through-plane electrical conductivity of the blends with the use of an Alpha-A high-performance conductivity analyzer (Novocontrol Technologies, Germany) at a frequency range of 10−1–105Hz and a voltage of 1V. We also used an ARES-G2 oscillation rheometer (TA Instruments, USA), equipped with parallel disks of 25mm diameter and 2mm spacing, to perform melt rheological measurements. Dynamic frequency tests were carried out at 200°C, under a continuous purge of dry nitrogen to avoid degrading the samples. Frequency sweeps between 0.1~100rad/s were performed at a strain amplitude of 10%, which is within the linear viscoelastic regime for all the blends we studied. We investigated the phase morphologies of the blends using a scanning electron microscope (JOEL 6060, 20kV). Samples were first cryofractured, and then the PS phase was preferentially extracted using tetrahydrofuran at room temperature. The etched samples were dried in a vacuum to remove any solvent, prior to gold coating.

Our results (Figure 1) show that electrical conductivity decreased with an increase in shear force and viscosity ratio. A faster screw speed, with stronger shear force, resulted in higher electrical conductivity. This result suggests that thermodynamics had a major effect on the selective localization of CNTs, while kinetics played a minor role. The blends prepared with the lowest melt viscosity ratio—see Figure 2(c) and 2(f)—showed larger co-continuous structures. As the screw speed and shear force increased, we observed that the blend structures became smaller, resulting in a refined morphology.


Electrical conductivity (lines) and shear force (bars) versus melt viscosity ratio for polypropylene/polystyrene (PP/PS) blends that contain carbon nanotubes (CNTs), with different melt flow rates for PS (10, 13, and 18g/10min, from left to right). The melt viscosity ratios of the blends are given in Table 2.


Scanning electron microscopy (SEM) images of the cryofractured surface of PP/PS blends prepared with CNTs (PP-CNTs/PS): (a) PP-CNTs/PS-10, (b) PP-CNTs/PS-13, and (c) PP-CNTs/PS-18 at 75rpm; and (d) PP-CNTs/PS-10, (e) PP-CNTs/PS-13, and (f) PP-CNTs/PS-18 at 100rpm. Scale bars indicate 10μm.

We also note that the blends prepared simultaneously had the highest electrical conductivity and the most uniform co-continuous structures (see Figures 3 and 4). This could be attributed to the structural differences that are induced by kinetically controlling particle location in the blend and the formation of percolating networks. We find that the changes in the screw speed, from 75 to 100rpm, had only a minor effect on morphology.


Electrical conductivity (lines) and shear force (bars) versus the sequence of incorporation for the PP-CNTs/PS-18 blends.


SEM images of the cryofractured surface of PP-CNTs/PS-18 blends with differing sequences of incorporation: (a) simultaneously, (b) PP-CNTs first, and (c) PS-18 first at 75rpm; and (d) simultaneously, (e) PP-CNTs first, and (f) PS-18 first at 100rpm. Scale bars indicate 10μm.


Electrical conductivity and shear force versus mixing time for the PP-CNTs/PS-18 blends.

The mixed thermodynamic-kinetic control of the CNT localization means that the high melt viscosity of the PS phase caused some of the CNTs to remain in the PP phase. As the mixing time increased, we were able to achieve better dispersion of CNTs in the polymer matrix, but with lower electrical conductivity. The reason for this may be that as mixing time increased, the polymer chains and CNTs were more likely to degrade and break. Moreover, the level of the migration and the percolation of the CNT network may also differ considerably, depending on the CNTs' shape or aspect ratio.

In summary, we have investigated a PP-CNTs/PS system in which thermodynamic factors were more influential than kinetic ones in the selective localization of CNTs. As a result, we were able to obtain a lower melt viscosity ratio, higher electrical conductivity, and a better co-continuous structure. The optimal compounding parameter for the PP-CNTs/PS-18 system was mixing the materials simultaneously, at 200°C within 2min. As a next step, we plan to study optimal conditions for annealing, including time and temperature, to minimize interfacial areas in the blends.


Authors

Kejing Yu
Jiangnan University

Kejing Yu's research interests include high-performance-fiber textiles, carbon-nanomaterial-filled lightweight polymer composite processing, and microcellular foaming. In 2014/2015 she was a visiting researcher at the Microcellular Plastics Manufacturing Laboratory at the University of Toronto, Canada, focusing on electrical conductive polymer composites.

Aboutaleb Ameli
Mechanical and Industrial Engineering, University of Toronto

Yasamin Kazemi
Mechanical and Industrial Engineering, University of Toronto

Sai Wang
Mechanical and Industrial Engineering, University of Toronto

Chul B. Park
Mechanical and Industrial Engineering, University of Toronto

Chul Park is a world leader in the development of innovative, cost-effective technologies for foamed plastics. He has been extensively involved in industrial projects (both in consulting and research contracts) on various foam processes, including microcellular processing, inert gas-injection processing, rotational foam molding, wood-fiber composites, and open-cell foams.


References

  1. K. Dai, X.-B. Xu and Z.-M. Li, Electrically conductive carbon black (CB) filled in situ microfibrillar poly(ethylene terephthalate) (PET)/polyethylene (PE) composite with a selective CB distribution, Polymer 48, pp. 849-859, 2007.

  2. A. Göldel, A. Marmur, G. R. Kasaliwal, P. Pötschke and G. Heinrich, Shape-dependent localization of carbon nanotubes and carbon black in an immiscible polymer blend during melt mixing, Macromolecules 44, pp. 6094-6102, 2011.

  3. M. Sumita, K. Sakata, S. Asai, K. Miyasaka and H. Nakagawa, Dispersion of fillers and the electrical conductivity of polymer blends filled with carbon black, Polym. Bull. 25, pp. 265-271, 1991.

  4. X. Zhao, J. Zhao, J.-P. Cao, D. Wang, G.-H. Hu, F. Chen and Z.-M. Dang, Effect of the selective localization of carbon nanotubes in polystyrene/poly(vinylidene fluoride) blends on their dielectric, thermal, and mechanical properties, Mater. Design 56, pp. 807-815, 2014.

  5. K. Yu, A. Ameli, Y. Kazemi, S. Wang and Chul. B. Park, The effect of matrix viscosity and compounding parameters on the morphology and electrical conductivity of PP-CNTs/PS blends, SPE ANTEC, 2015.

DOI:  10.2417/spepro.005831