Hyperbranched polysiloxane grafted graphene for improved tribological performance of bismaleimide composites

Abstract

Hyperbranched polysilane (HBPSi) grafted reduced graphene oxide (rGO), marked as HBPSi-rGO, was synthesized by the reaction of hydrosilylation. Fourier-transform infrared spectroscopy (FTIR) and X-ray photoelectron spectroscopy (XPS) indicated that the surface of graphene had been successfully functionalized. Subsequently, HBPSi-rGO was incorporated into the bismaleimides (BMI) resin to prepare composites so as to improve its performance. The tribological, mechanical and thermal properties of the composites were studied systematically. Results show that the addition of the appropriate content of HBPSi-rGO can enhance the mechanical properties including the impact and flexural strengths of the BMI resin. In addition, the thermal stability of HBPSi-rGO/BMI nanocomposites is also superior to that of the pure BMI resin. It is worth noting that when the content of HBPSi-rGO is 0.6 wt%, the frictional coefficient and the wear rate is decreased by 44.6% and 77.4%, respectively compared to those of the neat BMI. Scanning electron microscopy (SEM) reveals that wear mechanism of neat BMI is mainly fatigue wear, but it turns mainly adhesive wear after the incorporation of HBPSi-rGO. The main reason can be that the interface adhesion was enhanced due to the reaction of nucleophilic addition between HBPSi-rGO and BMI matrix.

---

1. Introduction

Bismaleimides (BMI) is a leading class of thermosetting polyimides based on low molecular weight building blocks and terminated by reactive groups, which undergo polymerization by thermal or catalytic means.¹ BMI resin is one of the most important thermosetting resins due to its excellent heat resistance, favorable electric insulation, good mechanical properties and dimensional stability, which enable its wide application in aerospace and electronic industries.² But the toughness of neat BMI resin is not satisfactory enough.³ Thus, in recent years, the modification of BMI resin has become a hot spot of research.^4,5

Graphene oxide (GO) is generally obtained by the oxidation of graphite via a modified Hummers method,^6–8 which can endow it with abundant carboxyl and carbonyl groups at the edges, whereas phenolic hydroxyl and epoxide functional groups are on the basal plane.^9–11 Because of the introduction of functional groups, dispersion of GO in most common solvents is improved significantly.^12–14 Hence, researchers' attention was attracted into the development of GO research. Yu et al.¹⁵ chemically grafted CH₂OH-terminated regioregular poly(3-hexylthiophene) (P3HT) onto the carboxylic groups of GO via an esterification reaction and the resultant P3HT-grafted GO sheets are soluble in common organic solvents. Zhang et al.¹⁶ modified GO with poly(N-vinylcarbazole), which exhibits an enhanced solubility in organic solvents. It is well known that reduced graphene oxide (rGO) is obtained from the reduction of GO by reducing agents, such as hydrazine hydrate, for which hydroxyl and carboxyl groups will be removed from surface of GO. Compared with GO, rGO is more advantageous when adopted as fillers into composites. First, esterification and hydrogen bonds can easily generate on the surface of GO for the existence of hydroxyl, carboxyl and epoxy groups, which would lead to the aggregation of GO. After reduction, aggregation of GO can be effectively prevented. Second, steric hindrance will drop after reduction of GO for the removal of hydroxyl and carboxyl groups, which is beneficial to further functionalize rGO.

Hyperbranched polymers have good solubility, low melt viscosity and a high density of functional groups, which makes them useful as surface modifiers.^17–19 Previously, we have utilized hyperbranched polysiloxane functionalized graphene oxide to modify cyanate ester, and high-performance cyanate ester was obtained successfully.^20–22 Thus, in this paper, hyperbranched polysiloxane was also chosen to functionalize rGO for its good thermal stability and high reactivity. Then, the functionalized rGO was incorporated into a BMI resin to improve its tribological and mechanical properties.

2. Experiment

2.1. Raw materials

BMI resin was supplied by Rongchang Ning research group at Northwestern Polytechnical University. Hyperbranched polysiloxane monomer (methylbis (dimethylallylsiloxy)silane) was prepared according to literature.²³ Vinyltriethoxysilane was purchased from Jianghan Fine Chemical LLC (Hubei, China). Hydrazine hydrate was purchased from Tianjin Fuchen Chemical Reagents Factory (Tianjin, China). Ammonia solution was purchased from Zhengzhou Paini Chemical Reagents Factory (Henan, China). Pt/C catalyst was purchased from Shanxi Kaida Chemical Engineering Co. Ltd (Tianjin, China). All chemicals were of analytical grade and used as received.

2.2. Preparation of HBPSi-rGO

2.2.1. Synthesis of vinyltriethoxysilane grafted graphene (VTES-G). GO was prepared from expandable graphite powder by a modified Hummers method and then HBPSi-rGO was prepared with GO and hyperbranched polysiloxane monomer via the reaction of hydrosilylation. GO (0.129 g) and deionized water (50 mL) were placed in a 250 mL three-neck flask and sonicated for 1 h. After adding vinyltriethoxysilane and HCl (1 mol L⁻¹), the mixture was stirred for 2 h at 75 °C. Then, hydrazine hydrate (2.5 mL) and ammonium hydroxide (2.5 mL) were used as reducing agents to deoxygenate GO at 95 °C for 4 h. After filtration, VTES-G was obtained (Fig. 1).

Fig. 1 Synthetic route of VTES-G.

2.2.2. Synthesis of HBPSi-rGO. VTES-G (0.1 g) was finely dispersed in CCl₄ (50 mL) by an ultrasonicator at room temperature. Hyperbranched polysiloxane monomers (0.5 g) and Pt/C catalyst (3 mg) were added with vigorous stirring under nitrogen atmosphere for 4 h at 50 °C. CCl₄ was removed from HBPSi-rGO through filtration, followed by washing with deionized water and ethylalcohol respectively for three times, as referenced (Fig. 2).²²

Fig. 2 Synthetic route of HBPSi-rGO.

2.2.3. Preparation of HBPSi-rGO/BMI composites. HBPSi-rGO/BMI composites were prepared by a casting method. BMI was modified by allyl phenyl ether in advance. First, HBPSi-rGO was finely dispersed in acetone by an ultrasonicator. The solid of BMI pre-polymer was melted into liquid at 120 °C in an oil bath. Then, HBPSi-rGO was added into the pre-polymer. The mixture was stirred vigorously for 30–45 min to evaporate acetone and obtain good homogeneity, and subsequently poured into a mould, which was preheated at 130 °C in an oven. The mixture in the mould was placed in a vacuum oven at 120 °C for 30–45 min to completely get rid of air and acetone bubbles. Finally, in an air dry oven, the mixture was cured at 130, 150, 180 °C for 1 h, and 200 °C for 2 h, and post-cured at 230 °C for 4 h. For comparison, the neat BMI specimen was fabricated in the same manner.

2.3. Property investigation of materials

2.3.1. X-Ray photoelectron spectroscopy. XPS spectra were recorded on a PHI Quantum 2000 Scanning ESCA Microprobe system.

2.3.2. Fourier-transform infrared spectroscopy. FTIR spectra were recorded on PerkinElmer-283B FT-IR Spectrometer. The samples were pressed into a pellet with KBr and scanned from 400–4000 cm⁻¹.

2.3.3. Friction and wear tests. The friction and wear tests were performed according to GB3960-83 (Chinese Standard) on a test machine (MM-200). The accuracy of the machine could achieve ±5% in measuring the frictional force and wear rate. Before each test, the counterpart steel ring and the materials were abraded with no. 900 water-abrasive paper. Then, the steel ring and samples were cleaned by acetone. All the friction and wear experiments were conducted at room temperature. The friction coefficient was measured under a load of 196 N and test duration of 120 min. The wear rate was measured using an electronic balance to an accuracy of ±0.0001 g.

2.3.4. Scanning electron microscopy (SEM) observation. The surface morphology of the wear surface of the samples was characterized with a CHIS-570 scanning electron microscope.

2.3.5. Mechanical properties. Impact strength and flexural strength were obtained according to GB/T2571-1981. The sample dimension was (50 ± 0.02) × (7 ± 0.02) × (4 ± 0.02) mm³. In each system, more than 5 specimens were tested.

2.3.6. Thermogravimetric analysis. The thermal stability was determined using TGAQ50 at a heating rate of 20 °C min⁻¹, in a nitrogen atmosphere.

3. Results and discussions

3.1. FTIR and XPS analysis of GO and HBPSi-rGO

Fourier Transform Infrared (FTIR) spectroscopy and X-ray photoelectron spectroscopy (XPS) were used for the investigation of elementary composition on the surface of GO and HBPSi-rGO.

Fig. 3 shows the FTIR spectras of GO and HBPSi-rGO. Three characteristic bands at 3427, 1620 and 1730 cm⁻¹ for GO correspond to –OH, C=C and C=O stretching vibrations in the hydroxyl, carboxyl, and skeleton, respectively. After chemical reduction and hydrosilylation addition, distinct absorption bands for HBPSi-rGO appeared at 1409, 1130 and 1030 cm⁻¹ corresponding to Si–C=C, Si–O–C and Si–O–Si, which indicate that hyperbranched polysilane has been grafted onto graphene. But C=O and –OH stretching vibrations were not detected because of chemical reduction. These results confirmed that GO and HBPSi-rGO have been successfully obtained.

Fig. 3 FTIR spectras of GO and HBPSi-rGO.

As shown in Fig. 4a, there are two intense peaks at 287.75 eV and 534.13 eV, which are attributed to the presence of the C and O element, respectively. Comparing the board scan XPS spectrum of HBPSi-rGO (Fig. 4b) with that of GO (Fig. 4a), the difference is the Si 1s peak at 153.89 eV and the Si 2p peak at 102.87 eV. Judging from the above evidence, it is convinced that GO and HBPSi-rGO was prepared successfully.

Fig. 4 Broad scan XPS spectras of (a) GO and (b) HBPSi-rGO.

Fig. 5 summarizes the results of the C 1s and Si 2p XPS spectras of HBPSi-rGO. According to Fig. 5a, the C 1s XPS spectra shows the most dominant peaks around 285.6 eV (C–C), accompanied by three small additional fitting-determined peaks around 284.4, 287.2 and 289.4 eV, which correspond to the following carbon components: C–Si, C–O and COOH. Fig. 5b shows two equivalent peaks in Si 2p spectra at 102.8 and 103.7 eV, which correspond to Si–O–C and Si–O–Si. There is a weak fitting peak at a higher binding energy (104.47 eV), which may be caused by dehydration of a little silanol to form silicone (*Si=O).

Fig. 5 XPS C 1s and Si 2p spectras and fitting curves of HBPSi-rGO.

By analyzing, the content of each element in HBPSi-rGO is presented in Table 1. The content of Si is 6.14 wt% as shown in Table 1 indicating that the quantity of hyperbranched polysiloxane grafted onto rGO is very satisfactory.

Table 1 Content of various elements in HBPSi-rGO

Element	Content/%
C	43.69
O	50.17
Si	6.14

---

There are many active functional groups on the surface and edge of GO, such as carboxyl, hydroxyl and epoxy groups. But these active groups will become an obstacle to a hyperbranched reaction for they can increase steric hindrance on basal planes. Thus, after modification with vinyltriethoxysilane, GO was reduced with hydrazine hydrate to increase the grafting ratio of HBPSi. In resin matrix, rGO acts as flexibilizer and lubricant for great mechanical stiffness and lubricity of graphene.²⁴

3.2. Mechanical properties of HBPSi-rGO/BMI composites

The impact strength (a) and flexural strength (b) of the composites with different contents of HBPSi-rGO are shown in Fig. 6. It can be seen that the impact strength and flexural strength of the composites are improved with an increase in HBPSi-rGO content from 0 to 0.6 wt%. When the HBPSi-rGO content is 0.6 wt%, the impact strength and flexural strength respectively reach the peaks (18.81 kJ m⁻² and 150.52 MPa), and increase as much as 62.6% and 20.2% in comparison with the neat BMI resin. This can be attributed to the following reasons: (1) the good dispersion of HBPSi-rGO in BMI matrix, and 2D-lamellar structure of HBPSi-rGO nanosheets, which will benefit dispersion of stress, (2) the nucleophilic addition can enhance the interfacial bonding between double bonds of HBPSi-rGO and BMI resin (as shown in Fig. 7),²⁵ which will prevent the fracture of the composite.

Fig. 6 Impact strength (a) and flexural strength (b) of composites with different contents of HBPSi-rGO.

Fig. 7 Nucleophilic addition between double bonds of HBPSi-rGO and BMI resin.

However, the impact strength and flexural strength of the composites decrease when the HBPSi-rGO content is 0.6 wt%. A possible reason for this decrease can be due to the aggregation of HBPSi-rGO at high concentration.

3.3. Thermal properties of HBPSi-rGO/BMI composites

To investigate the influence of HBPSi-rGO on the thermal properties of BMI, Thermogravimetric Analysis (TGA) was carried out to test the neat BMI and composites with 0.6 wt% HBPSi-rGO (as shown in Fig. 8). The initial thermal decomposition temperature of neat BMI and HBPSi-rGO/BMI composites is approximately the same, and both are high. But the carbon yield of HBPSi-rGO/BMI composites at 800 °C is 4.7% higher than that of the neat BMI. The carbon yield for HBPSi-rGO/BMI is 29.5% and neat BMI is 24.8%. This can be attributed to the reaction of nucleophilic addition between the HBPSi-rGO and BMI resin, which might improve the crosslink and prevent polymer chains from decomposition at high temperature.

Fig. 8 TG curves of neat BMI and 0.6 wt% HBPSi-rGO/BMI composite.

3.4. Tribological properties of HBPSi-rGO/BMI composites

Fig. 9 shows friction coefficient (a) and wear rate (b) values of the HBPSi-rGO/BMI composites with different contents of HBPSi-rGO. It can be found that when the addition of HBPSi-rGO is below 0.6 wt%, the friction coefficient and wear rate drop gradually with increase in the content of HBPSi-rGO. When the HBPSi-rGO content is 0.6 wt%, the coefficient and wear rate both reach the lowest, and are 0.194 and 1.4 × 10⁻⁶ mm³ N⁻¹ m⁻¹, respectively. The significant decrease in the friction coefficient and wear rate for the composites can be associated with the following three reasons. First, during the wear process, the flake structure of HBPSi-rGO nanosheets makes the composites easily form transfer films on the surface of steel counterpart. This can prevent direct contact between asperities on the steel counterpart and the surface of the composite block, thereby improving the friction properties of the composites. Second, in the process of friction, tremendous heat will generate on the friction surface, where the structure of composites may be destroyed at high-temperature. However, friction heat in the composites can be transmitted efficiently because of good thermal conduction of HBPSi-rGO, resulting in the inhibition of destruction caused by high-temperature.²⁶ In addition, the improvements in mechanical properties can enhance loading capacity of composites during a frictional test.

Fig. 9 Frictional coefficient (a) and wear rate (b) of composites with different content HBPSi-rGO.

However, when the incorporation of HBPSi-rGO exceeds 0.6 wt%, the friction coefficient maintains the lowest level, and wear rate dramatically increases, but both are still much lower than neat BMI matrix. This can be attributed to aggregation of HBPSi-rGO nanosheets at high contents, which results in stress concentrations at HBPSi-rGO aggregate sites, and then leads to a higher wear rate. But all in all, the incorporation of HBPSi-rGO at a very low content into the BMI matrix has led to significant improvement in tribological performance.

Fig. 10 shows SEM images of the wear surface of neat BMI and 0.6 wt% HBPSi-rGO/BMI under the same testing condition. Cracking signs covering the whole worn surface of the neat BMI can be seen in Fig. 10a. Fig. 10b shows smaller and milder cracking signs when the incorporation of HBPSi-rGO is 0.6 wt%. All these observations indicate that the anti-wear ability of composites is reinforced by HBPSi-rGO.

Fig. 10 SEM images of wear surface of (a) neat BMI, (b) 0.6 wt% HBPSi-rGO/BMI composites.

Fig. 11 shows the transfer film of neat BMI and composites with 0.6 wt% HBPSi-rGO. As shown in Fig. 11a, a lot of notches were found on counter surface for neat BMI, which indicates that uniform and tenacious transfer film cannot form for the neat BMI composite under dry sliding conditions. Therefore, the neat BMI presents a high frictional coefficient and a weak wear resistance. But as shown in Fig. 11b, when the HBPSi-rGO content reaches 0.6 wt%, the worn surface becomes smoother and obvious notches disappear. These results of SEM research correspond to the low frictional coefficient and wear rate of the BMI composite with 0.6 wt% HBPSi-rGO (Fig. 9).

Fig. 11 Microimages of the counter surface of (a) neat BMI, (b) 0.6 wt% HBPSi-rGO/BMI composites.

4. Conclusions

A kind of high-performance polymer composite has been fabricated using BMI resin as a matrix and rGO as filler by casting method, in which the rGO was grafted by hyperbranched polysiloxane. The tribological and mechanical performances of BMI resin were greatly improved by the addition of HBPSi-rGO at a low content. These can be attributed to the excellent properties of graphene sheets, such as lubricity and thermal conductivity. In addition, the reaction between HBPSi-rGO and the BMI matrix can enhance the interfacial bonding and compatibility of HBPSi-rGO with the BMI matrix.

Acknowledgements

This work was financially supported by the Doctorate Foundation of Northwestern Polytechnical University (CX201429) and the Research Fund for the Doctoral Program of Higher Education (20136102110049).

References

1. A. Gu, Compos. Sci. Technol., 2006, 66, 1749.
2. V. Rao, S. Navath, M. Kottur, J. R. McElhanon and D. V. McGrath, Tetrahedron Lett., 2013, 54, 5011.
3. R. J. Morgan, E. E. Shin, B. Rosenberg and A. Jurek, Polymer, 1997, 38, 639.
4. H. X. Yan, Y. Jia, M. L. Li and T. T. Li, J. Appl. Polym. Sci., 2013, 129, 3150.
5. Y. Jia, H. X. Yan, L. Ma and J. P. Zhang, J. Polym. Res., 2014, 21, 1.
6. Y. W. Zhu, S. Murali, W. W. Cai, X. S. Li, J. W. Suk, J. R. Potts and R. S. Ruoff, Adv. Mater., 2010, 22, 3906.
7. D. C. Marcano, D. V. Kosynkin, J. M. Berlin, A. Sinitskii, Z. Z. Sun, A. Slesarev, L. B. Alemany, W. Lu and J. M. Tour, ACS Nano, 2010, 4, 4806.
8. S. Stankovich, D. A. Dikin, G. H. B. Dommett, K. M. Kohlhaas, E. J. Zimney, E. A. Stach, R. D. Piner, S. T. Nguyen and R. S. Ruoff, Nature, 2006, 442, 282.
9. C. Gómez-Navarro, R. T. Weitz, A. M. Bittner, M. Scolari, A. Mews, M. Burghard and K. Kern, Nano Lett., 2007, 7, 3499.
10. D. R. Dreyer, S. Park, C. W. Bielawski and R. S. Ruoff, Chem. Soc. Rev., 2010, 39, 228.
11. A. Bagri, C. Mattevi, M. Acik, Y. J. Chabal, M. Chhowalla and V. B. Shenoy, Nat. Chem., 2010, 2, 581.
12. J. I. Paredes, S. Villar-Rodil, A. Martínez-Alonso and J. M. D. Tascón, Langmuir, 2008, 24, 10560.
13. D. Li, M. B. Müller, S. Gilje, R. B. Kaner and G. G. Wallace, Nat. Nanotechnol., 2008, 3, 101.
14. Y. P. Wu, B. Wang, Y. F. Ma, Y. Huang, N. Li, F. Zhang and Y. S. Chen, Nano Res., 2010, 3, 661.
15. D. S. Yu, Y. Yang, M. Durstock, J. Baek and L. M. Dai, ACS Nano, 2010, 4, 5633.
16. B. Zhang, Y. Chen, X. D. Zhuang, G. Liu, B. Yu, E. Kang, J. H. Zhu and Y. X. Li, J. Polym. Sci., Part A: Polym. Chem., 2010, 48, 2642.
17. T. Ganicz, T. Pakula, W. Fortuniak and E. Białecka-Florjanczyk, Polymer, 2005, 46, 11380.
18. E. B. Tarabukina, A. A. Shpyrkov, E. V. Tarasova, A. I. Amirova, A. P. Filippov, N. A. Sheremet'eva and A. M. Muzafarov, Polym. Sci., Ser. A, 2009, 51, 150.
19. Y. Zheng, K. J. Thurecht and W. X. Wang, J. Polym. Sci., Part A: Polym. Chem., 2012, 50, 629.
20. M. M. Zhang, H. X. Yan, C. Gong and T. T. Li, eXPRESS Polymer Letters, 2014, 8, 413.
21. M. M. Zhang, H. X. Yan, C. Gong and F. F. Zhang, J. Compos. Mater., 2014 DOI:10.1177/0021998314527778.
22. M. M. Zhang, H. X. Yan, X. W. Yang and C. Liu, RSC Adv., 2014, 4, 45930.
23. Q. F. Si, X. Wang, X. D. Fan and S. J. Wang, J. Polym. Sci., Part A: Polym. Chem., 2005, 43, 1883.
24. S. Stankovich, D. A. Dikin, G. H. B. Dommett, K. M. Kohlhaas, E. J. Zimney, E. A. Stach, R. D. Piner, S. T. Nguyen and R. S. Ruoff, Nature, 2006, 442, 282.
25. A. J. Gu, G. Z. Liang and L. W. Lan, Thermosetting Resin, 1995, 10, 38 , (in Chinese).
26. B. L. Pan, S. P. Zhang, W. Z. Li, J. Zhao, J. L. Liu, Y. Q. Zhang and Y. Z. Zhang, Wear, 2012, 294, 395.

---