Root-like glass fiber with branched fiber prepared via molecular self-assembly

Abstract

This paper developed a mild and controlled strategy to build in situ a root-like glass fiber (GF) via molecular self-assembly in polypropylene melts to achieve the interfacial mechanical interlocking and facilitate load transfer from the polymer matrix to the reinforcing fiber, which is very important for getting the utmost out of GF's reinforcement.

---

Fiber-reinforced polymer composites are widely used in advanced aerospace systems and automotive industries.^1,2 Generally, the loads are transferred from the polymer matrix to the reinforcing fiber via their interface, which governs the reinforcement efficiency of the fiber and the final mechanical performances.³ However, current composites often cannot present full reinforced potential compared to their theoretic prototype because of weak interfacial interaction between the polymer matrix and reinforcing fiber.^4,5 Therefore, interfacial design and optimization are the long-standing critical issues for fiber-reinforced polymer composites and many efforts have been devoted to guarantee good interfacial interaction and prepare excellent-performance composite products. Even where the interfacial interaction still isn't strong enough to ensure the effective load transfer, the reinforcing fiber can be pulled out simply from the polymer matrix.⁶

Recently, inspired by the geometrical feature of tree root with mechanical interlocking effect with the surrounding soil which is pulled out difficultly, some branched reinforcing fiber were designed to improve the interfacial interaction.^7–9 Depending on the density, dimension and alignment of the branched fiber as well as its compatibility with polymer, the interfacial shear strength of the branched fiber reinforced polymer composites can increase by 10–150%.^10–13 Tamrakar adopted electrophoretic deposition method to achieve the growth of carbon nanotubes (CNTs) on the surface of glass fibers (GF) and obtained CNTs-g-GF/epoxy composites with 58% increase in interfacial shear strength (IFSS).¹¹ Sager showed that IFSS of epoxy composites containing CNTs-g-carbon fiber with radially-aligned orientations was 54% higher than that of the composites with random fiber.¹⁴ However, main routes to deposit the branched fiber on primary fibers still presented some unsolved issues. For chemical vapor deposition, high temperature and predeposited catalysts can deteriorate the strength of primary fibers. Sizing or chemical reactions can't tailor the alignment of branched fibers, which often lie at the surface of the primary fiber. What's more, the root-like fibers are only applied in the thermoset polymer because of the self-filtration and high viscosities during the blending process of the reinforcing fibers with polymer matrix.^15–17 Therefore, there is an urgent need for developing a mild strategy suitable for thermoplastic polymer to prepare the branched reinforcing fiber. This paper achieved in situ epitaxial growth of aryl amide-based fiber on the surface of GF in polypropylene (PP) melts by tailoring the molecular self-assembly and its wetting behaviors on GF to prepare the root-like GF having mechanical interlocking with polymer matrix, which can potentially improve the interfacial interaction and the mechanical properties of the GF/PP composites.

TMB-5, a well-known aryl amide-based nucleating agent for PP, can dissolve in PP melts at high temperature and self-assemble into fibrous structure via intermolecular hydrogen bonding upon cooling.^18,19 Polarized light microscope (PLM) was adopted to investigate the dissolution and self-assembly behaviours during the heating and subsequent cooling processes and the typical POM photos were shown in Fig. 1. Clearly, during the heating of PP/TMB-5 blend, TMB-5 gradually dissolved into the PP melts, followed by the formation of the dendrites from the homogeneous melt in the sequent cooling stage.

Fig. 1 PLM photos of PP with 0.2% TMB-5 during the heating and cooling process at a rate of 30 °C min⁻¹: (a) 180 °C; (b) 210 °C; (c) 240 °C; (d) 260 °C; (e) 190 °C; (f) 180 °C; (g) 150 °C; (h) 145 °C; (i) 140 °C.

It's evident that the self-assembly process is nucleation-dominated.^20,21 According to the thermodynamic lowest energy principle, heterogeneous nucleation is more favourable than homogeneous nucleation.²² If GF can serve as the nucleating site for the self-assembly of TMB-5, the latter will preferably grow on the surface of the former, forming the root-like GF with branched TMB-5 fiber. To this end, GF firstly was insert into the PP containing 0.2% TMB-5 at 180 °C and then heated to 260 °C at a rate of 30 °C min⁻¹, where TMB-5 can dissolve completely into PP melts, beneficial to observing the self-assembly of TMB-5. As shown in Fig. 2a, neat GF failed to induce the epitaxial growth of TMB-5, which separately self-assembled into the dendrites. This is ascribed to the big difference of their surface natures: the surface energy of GF was 26.1 mN m⁻¹ while that of TMB-5 was 39.6 mN m⁻¹. Thus, TMB-5 can't tend to diffuse onto the surface of GF. In order to improve the compatibility of GF with TMB-5, GF was dispersed into dopamine solution and then polydopamine layers were coated onto the surface of GF through the oxidation of dopamine at room temperature²³ [SEM photos of neat GF and polydopamine coated GF are provided in ESI†]. Since massive OH groups was introduced into the surface of GF due to the existence of polydopamine, the surface energy of the modified GF became 38.6 mN m⁻¹, similar to that of TMB-5; moreover, there were multi-hydrogen bond interactions between OH groups of the modified GF and NH groups of TMB-5 molecules. The two were beneficial to the wetting and preferential deposition of TMB-5 molecules on the surface of GF. Therefore, the modified GF can functions as the nucleating template to direct the radial growth of TMB-5 fiber on the GF, forming the root-like GF with laterally grown TMB-5 fiber (Fig. 2b). Accordingly, the formation mechanism of root-like structure was proposed as shown in Fig. 2c. Similar to the tree root, TMB-5 fiber at the PP/GF interface can act as the intermediate bridge to make GF mechanical interlock with the surrounding polymer to enhance the difficulty of its pull-out from polymer matrix under mechanical load, impacting strong interfacial interaction to the composites. The load can transfer more effectively from PP matrix to the GF, which is crucial for getting the utmost out of GF's reinforcement. Additionally, it should be mentioned that TMB-5 is a good nucleating agent for PP and thus polymer crystals tend to grow on the surface of the branched TMB-5 fiber, which is verified by Fig. 2b2. This densely transcrystalline can enhance the interfacial interaction between polymer and the branched fiber,²⁴ which will improve the interlocking effect of the root-like GF with PP matrix and fix GF stronger in polymer matrix.

Fig. 2 PLM photos for high-viscosity PP (melt index (MI) = 0.5) with 0.2% TMB cooled from 260 °C to 180 °C at a rate of 10 °C min⁻¹: (a) neat GF; (b) polydopamine coated GF; (c) the proposed formation mechanism of root-like GF with branched TMB-5 fiber.

Based on the formation and growth of the root-like GF, two factors, heterogeneous nucleation of TMB-5 on GF and the diffusion rate in PP melts, play important roles in the density and size of the branched TMB-5 fiber. Fig. 3 showed the effects of the cooling rate and viscosity of PP melts. With the increasing cooling rate, the self-assembly of TMB-5 molecules preferred to proceed on the surface of the affinitive GF. Therefore, the number of the branched TMB-5 fiber increased and more nucleation sites also decreased the lateral dimension from 733 μm to 163 μm, as shown in Fig. 2b and 3b. On the other hand, TMB-5 molecules moved more easily in low viscosity PP melts so that at the same cooling rate, the growth of the branched fiber was more perfect and the root-like GF finally exhibited bigger dimension (Fig. 3c) compared to that in the high-viscosity PP melts (Fig. 3b). Obviously, the density and dimension of branched TMB-5 fiber can be finely tailored by adjusting the cooling rate and viscosity of PP melts, providing a simple and facile way to achieve the controlled self-assembly of TMB-5 on the surface of GF and microstructure optimization of the root-like GF.

Fig. 3 PLM photos for high viscosity PP (MI = 0.5) with 0.2% TMB cooled from 260 °C at different cooling rate: (a) 20 °C min⁻¹, (b) 30 °C min⁻¹; (c) low viscosity PP (MI = 12) with 0.2% TMB at the cooling rate of 30 °C min⁻¹.

Finally, based on the above preparation strategy, 10 wt% polydopamine-coated GF and PP containing 0.2% TMB were blended and then injection-moulded on HAAKE MiniJet Piston Injection machine. As expected, the root-like GF also was generated in the prepared GF/PP composite during the practice processing and the yield strength reached 54.0 MPa, higher than 43.0 MPa of neat GF/PP composite, which confirmed that the root-like GF can get the more utmost out of GF's reinforcement [PLM of root-like GF and stress–stain curves of the two GF/PP composites are provided in ESI†].

In summary, the novel root-like fiber composed of glass fiber as stem and nucleating agents for PP as branch was prepared via self-assembly of TMB-5 on the surface of GF. Expectedly, the root-like fiber with densely transcrystalline layer of PP can mechanically interlock with the surrounding polymer matrix to maximize the interfacial load transfer and the reinforcement efficiency of GF, solving the interfacial problems of fiber-reinforced polymer composites.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (51127003, 51303114 and 51421061).

Notes and references

1. J. Karger-Kocsis, H. Mahmood and A. Pegoretti, Prog. Mater. Sci., 2015, 73, 1–43.
2. M. Sharma, S. Gao, E. Mäder, H. Sharma, L. Y. Wei and J. Bijwe, Compos. Sci. Technol., 2014, 102, 35–50.
3. F. Gardea, B. Glaz, J. Riddick, D. C. Lagoudas and M. Naraghi, ACS Appl. Mater. Interfaces, 2015, 7, 9725–9735.
4. L. Tzounis, S. Debnath, S. Rooj, D. Fischer, E. Mäder, A. Das, M. Stamm and G. Heinrich, Mater. Des., 2014, 58, 1–11.
5. L. Tzounis, M. Kirsten, F. Simon, E. Mäder and M. Stamm, Carbon, 2014, 73, 310–324.
6. C. Cui, W. Qian, M. Zhao, F. Ding, X. Jia and F. Wei, Carbon, 2013, 60, 102–108.
7. K. Liu, M. Jin, R. La, J. Zhang, T. Wang and X. Zhang, Mater. Lett., 2014, 125, 209–212.
8. F. Meng, R. Zhao, Y. Zhan and X. Liu, J. Mater. Chem., 2011, 21, 16385–16390.
9. H. Qian, G. Kalinka, K. L. A. Chan, S. G. Kazarian, E. S. Greenhalgh, A. Bismarck and M. S. P. Shaffer, Nanoscale, 2011, 3, 4759–4767.
10. Y. Li, Y. Li, Y. Ding, Q. Peng, C. Wang, R. Wang, T. Sritharan, X. He and S. Du, Compos. Sci. Technol., 2013, 85, 36–42.
11. S. Tamrakar, Q. An, E. T. Thostenson, A. N. Rider, B. Z. Haque and J. W. Gillespie, ACS Appl. Mater. Interfaces, 2016, 8, 1501–1510.
12. C. Wang, X. He, L. Tong, Q. Peng, R. Wang, Y. Li and Y. Li, Composites, Part A, 2013, 50, 1–10.
13. C. Wang, Y. Li, L. Tong, Q. Song, K. Li, J. Li, Q. Peng, X. He, R. Wang, W. Jiao and S. Du, Carbon, 2014, 69, 239–246.
14. R. J. Sager, P. J. Klein, D. C. Lagoudas, Q. Zhang, J. Liu, L. Dai and J. W. Baur, Compos. Sci. Technol., 2009, 69, 898–904.
15. Q. An, A. N. Rider and E. T. Thostenson, ACS Appl. Mater. Interfaces, 2013, 5, 2022–2032.
16. E. Bekyarova, E. T. Thostenson, A. Yu, H. Kim, J. Gao, J. Tang, H. T. Hahn, T. W. Chou, M. E. Itkis and R. C. Haddon, Langmuir, 2007, 23, 3970–3974.
17. H. Qian, E. S. Greenhalgh, M. S. P. Shaffer and A. Bismarck, J. Mater. Chem., 2010, 20, 4751–4762.
18. M. Dong, Z.-X. Guo, J. Yu and Z.-Q. Su, J. Polym. Sci., Part B: Polym. Phys., 2009, 47, 314–325.
19. R. Han, Y. Li, Q. Wang and M. Nie, RSC Adv., 2014, 4, 65035–65043.
20. F. Luo, K. Wang, N. Ning, C. Geng, H. Deng, F. Chen, Q. Fu, Y. Qian and D. Zheng, Polym. Adv. Technol., 2011, 22, 2044–2054.
21. Q. Kong, Q. Liao, Z. Xu, X. Wang, J. Yao and H. Fu, J. Am. Chem. Soc., 2014, 136, 2382–2388.
22. X. Y. Liu, J. Chem. Phys., 2000, 112, 9949–9955.
23. B. Fei, B. Qian, Z. Yang, R. Wang, W. C. Liu, C. L. Mak and J. H. Xin, Carbon, 2008, 46, 1795–1797.
24. J. P. Abdou, G. A. Braggin, Y. Luo, A. R. Stevenson, D. Chun and S. Zhang, ACS Appl. Mater. Interfaces, 2015, 7, 13620–13626.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra07240b

---