Effect of surface-functionalized reduced graphene oxide on mechanical and tribological properties of bismaleimide composites

Abstract

Surface-functionalized reduced graphene oxide sheets were obtained by grafting hyperbranched polytriazine (HBPT) onto the surface of reduced graphene oxide (RGO) sheets. The HBPT was prepared through repeating nucleophilic substitution between chloride ions of cyanuric chloride and amino groups of hexamethylenediamine. The hyperbranched polytriazine modified reduced graphene oxide (HBPT-RGO) was then added as the nanofiller and polyamine curing agent in diallyl bisphenol A modified bismaleimide resin (BMI-BA) to fabricate in situ the HBPT-RGO/BMI-BA nanocomposite. The characterization results showed that suitable addition of HBPT-RGO could greatly enhance the mechanical properties and decrease the frictional coefficient and wear rate of the BMI-BA resin dramatically, which might be related to the good compatibility and interfacial adhesion between HBPT-RGO and the BMI-BA matrix.

---

1. Introduction

Thermosetting bismaleimide (BMI) resin has evolved as one of the most important matrix materials for advanced polymer composites because of their “epoxy-like” processing characteristics, low evolution of volatile gases, high glass transition temperature (T_g), high modulus, low water absorption, good retention of mechanical properties at high temperatures, and superior thermal stability and fatigue resistance under conditions of high humidity.^1–3 Many of their impressive properties, including high glass transition temperature and high modulus, are directly related to the underlying microstructure of high crosslinking density and rigid molecular network. However, this structural feature also results in an inherent brittleness of the material.^4,5 In the past two decades, different approaches to toughen bismaleimide had been investigated.^6–9 The addition of organic and inorganic fillers or particles to a polymer has been a way to improve the properties of materials. The addition of fillers into the matrix can usually enhance load withstanding capability, reduce coefficient of friction, and improve electrical, mechanical and thermal properties.^10–12 The popular studied approach to toughen thermosetting resin is to add nanoparticles as the modifier. This approach can not only improve the fracture toughness, but also bring significant increase in many desirable properties such as glass transition temperature and yield strength.^13,14

Graphene, the strictly two dimensional layers of sp²-bonded carbon, is now a rising star in materials science. Its extraordinary electronic transport property, excellent thermal conductivity, mechanical performance, low density, high surface area and self-lubrication offer great promise for many practical applications including electronics, sensors and actuators, solar cells and data storage, optics and photonics, medical and biological applications, tissue engineering and biomaterials, and functional nanocomposites.^15–17 Fu et al. reported that the presence of graphene oxide (GO) and its hybrid could greatly increase tribological performance of epoxy nanocomposites.^18,19 Compared to graphene oxide, the graphene has better thermal stability and thermal conductivity. This benefits heat transfer and improving the thermal stability of graphene-based composites during the wear process.²⁰ Many studies have demonstrated that graphene filled polymer composites showed enhanced tribological property with friction coefficient and wear rate reduced significantly, when compared to the original polymer composites.^21,22 However, the properties of graphene-based polymer composites are strongly influenced by the aggregation, restacking of graphene and weak interaction between graphene and polymeric matrixes.^23,24 Thus, these may result in failing to achieve the desired properties and even unavailable for the properties of graphene-based polymer composites. The functionalized graphene sheets can solve these problems mentioned above by providing multiple bonding sites on graphene sheets with the polymer matrix. The interface interaction between the nanoparticles and polymer matrix in the nanocomposites, which is conducive to stress transfer.^25,26 Recently, hyperbranched polymers have attracted increasing attention because of their interesting architecture and unique physicochemical properties.^27,28 Hyperbranched polymers not only have a highly branched, non-entangled architecture and a large number of terminal groups but also exhibit much lower melt and solution viscosities and higher solubility in comparison with their linear analogues.^29,30 We have reported that the cyclotriphosphazene polymers was successfully grafted onto reduced graphene oxide (HBP-RGO) surface to form a stable dispersion by the repeating nucleophilic substitution of hexachlorocyclotriphosphazene and hexamethylenediamine.³¹ Nevertheless, a high degree of cross-linking and expensive raw materials affect the intrinsic properties of graphene and hinder its wide application in composite materials.

Therefore, the hexachlorocyclotriphosphazene was replaced by cyanuric chloride to prepare the hyperbranched polytriazine modified reduced graphene oxide, which had lower steric hindrance, cheaper and better applicability. In this article, the HBPT-RGO was added as a nanofiller and polyamine curing agent in BMI-BA resin and then synthesis in situ HBPT-RGO/BMI-BA composites by casting method. The HBPT-RGO was well dispersed in the BMI-BA composites and display good interfacial adhesion with the BMI-BA. The HBPT-RGO/BMI composites presented excellent mechanical and tribological properties. This investigation is aiming to prepare a new kind of composites with excellent mechanical and tribological properties.

2. Experimental

2.1. Reagents and materials

Natural graphite flakes (500 mesh) was obtained from Qingdao Hensen Graphite Co., Ltd. The graphene oxide (GO) nanosheets were produced from natural graphite flakes by the modified Hummers method.³² γ-Aminopropyltriethoxysilane (KH-550) was purchased from Jingzhou Jianghan Fine Chemical Co., Ltd. Cyanuric chloride (CC, 99%), hexamethylene-diamine (HMD) and triethylamine (TEA) were purchased from Aladdin Chemistry Co., Ltd. Tetrahydrofuran (THF), diethyl ether and methanol were purchased from Tianjin Tianda Chemical Co. Ltd. The BMI was provided by Rongchang Ning research group at Northwestern Polytechnical University. Diallyl Bisphenol A (DBA) was purchased from Sigma-Aldrich. All reagents were of analytical grade and used as received without further purification.

2.2. Preparation

Introduction of amino groups onto the reduced graphene oxide surface. GO was prepared by a modified Hummers method.³² Firstly, H₂O (7 mL), ethanol (75 mL) and GO (0.2 g) were added into a three-necked flask equipped with a purge for N₂ gas, and a reflux condenser. After the atmosphere of it was replaced with N₂ gas. The γ-aminopropyltriethoxysilane (KH-550) ethanolic solution (0.3 wt%) was added in at 78 °C under agitation at 400 rpm and kept for 2 h to produce the functionalized GO, the product was filtered and washed by ammonia and ethanol. The surface amino-functionalized graphene oxide was abbreviated GO-NH₂. Secondly, the GO-NH₂ obtained from the first step, hydrazine hydrate (2 mL) and H₂O (200 mL) were added into a three-necked flask at 98 °C kept for 4 h, as ref. 31. Finally, the solution was filtered and washed by deionized water and ethanol. The final products were stored in a vacuum oven at 60 °C until constant weight. The surface amino-functionalized reduced graphene oxide was abbreviated G-NH₂. The reduction of graphene oxide (RGO) was prepared in similar procedures.

Grafting of hyperbranched polytriazine onto the G-NH₂. The first step was carried out as follows: G-NH₂ (0.4 g), cyanuric chloride (0.92 g), NaOH (0.2 g), triethylamine (1 mL) and tetrahydrofuran (70 mL) were mixed in a 250 mL round bottom flask at ice water bath for 6 h. After the reaction, the resultant was filtrated and washed by ethanol to remove the unreacted reagents. The second reaction step was carried out as follows: the resultant obtained from the first step, hexamethylenediamine (1.16 g) NaOH (0.4 g), triethylamine (2 mL) and tetrahydrofuran (70 mL) were charged into a three-necked, round-bottom flask equipped with an addition mechanical stirrer, reflux condenser and constant-pressure funnel. The mixture was kept stirring at 60 °C for 4 h. The system was then heated to 90 °C for 3 h. The final products were filtrated and washed by ethanol until no more unreacted monomer or by-products could be detected in the filtrate. Both the first and second reaction steps were repeated four times to grow the hyperbranched polytriazine onto graphene surfaces (each generation the amount of cyanuric chloride and hexamethylenediamine was added at a molar ratio of 1 : 2).¹⁰ After the above procedures, the resulting fluffy black sample (see Fig. S5†) was dried in a vacuum oven at 60 °C. The hyperbranched polytriazine modified reduced graphene oxide was abbreviated HBPT-RGO. The chemical route for the preparation of HBPT-RGO was illustrated in Fig. S1.†

Preparation of HBPT-RGO/BMI-BA composites. The HBPT-RGO/BMI-BA composites were prepared by casting method. Diallyl Bisphenol A (DBA) and BMI (DBA and BMI with a mass ratio of 3 : 4) as well as the HBPT-RGO were mixed under vigorous stirring conditions at 135 °C (200 rpm) till the BMI-BA totally melting and the HBPT-RGO was uniformly dispersed in the mixture. Afterward, the mixture was poured into a pre-heated mold with release agent, and degassed in a vacuum drying oven at 150 °C for 0.5–1 h. Finally, the mixture was cured following the schedule of 150 °C/2 h + 180 °C/2 h + 220 °C/4 h. A post curing process was 250 °C/4 h.

2.3. Characterizations

The synthesized product was characterized by X-ray diffraction (XRD, Rigaku, model D/max-2500 system at 40 kV and 100 mA of Cu Kα). The friction and wear tests were performed according to GB3960-83 (Chinese Standard) on a test machine (MM-200, load 198 N, speed 400 rpm) under dry condition. Raman spectra was obtained using Thermo Nicolet Ahmega Dispersive Raman with a Nd:YAG laser and liquid N₂ cooled Ge detector with laser excitation at 672 nm. X-ray photoelectron spectra (XPS) were recorded by a PHI Quantum 2000 Scanning ESCA Microprobe system. All XPS spectra were corrected using the C1s line at 285.0 eV. Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images were obtained via a FEI Nova Nano SEM 230 and a FEI Tecnai G2 F20 microscopy, respectively. Impact strength and flexural strength were obtained according to GB/T2567-2008. The samples dimension for impact test and flexural test were (80 ± 0.2) × (10 ± 0.2) × (4.0 ± 0.2) and (80 ± 0.2) × (15 ± 0.2) × (4.0 ± 0.2) mm³, respectively. The thermal stability was determined using TGA Q50 at a heating rate of 10 °C min⁻¹, in a nitrogen atmosphere.

3. Results and discussion

3.1. The characterizations of HBPT-RGO

In Fig. 1a, the peaks at 102.3, 285.1, 399.6 and 532.0 eV were attributed to Si2p, C1s, N1s and O1s, respectively. As presented in Fig. 1b, the four peaks emerging at 284.5, 286.6, 287.5, and 288.7 eV were ascribed to C–C, C–OH, C–O–C, and C=O bonds, respectively.³³ In Fig. 1c, the Si2p just could be fitted to one fitting curve belonged to Si–O–C, which confirmed that the KH-550 of self-aggregation after hydrolyze had been basically cleared. In Fig. 1d, the C1s peak of HBPT-RGO could be fitted to five fitting curves with the binding energy located at 284.5, 285.6, 287.5, 288.1 and 288.7 eV, which belonged to C–C, C–N, C–O–C, C=N and C=O, respectively. Compared to C1s XPS spectrum of GO, the presence of an additional peak of C1s at 288.1 eV (C=N) in Fig. 1d confirmed that the cyanuric chloride was successfully grafted onto graphene nanosheets via substitution reaction.³⁴ These results are in good agreement with the results of FT-IR (see Fig. S2†).

Fig. 1 (a) XPS spectra of all the samples; (b) C1s of GO; (c) Si2p of G-NH₂; (d) C1s of HBPT-RGO.

XRD was performed to detect the variations of the layer structure among the samples and the patterns were depicted in Fig. 2. After oxidation, the d-spacing of graphite sheets increased from 0.336 nm to 0.786 nm, indicating that the inter-layer distance increased due to the various oxygen-containing groups (the hydroxyl, carboxyl and epoxy groups).³⁵ The characteristic peak of graphite at 26.5° disappeared in GO, and a new peak that belonged to GO occurred at 11.2°, which indicated that graphite had been fully oxidized.³⁶ After grafting the KH-550, the peak of GO moved to 9.23°, which was attributed to the KH-550 grafting onto the GO surface and thereby the d-spacing of GO increased from 0.786 to 0.957 nm. There was an obvious XRD diffraction peak at 24.3°, which was assigned to graphene. This was indicated that the graphene oxide sheets had been reduced and stripped.³⁷ After grafting the hyperbranched polytriazine, the HBPT-RGO presented a wide and low peak at 21.3°, indicating the interlayer spacing of the sheets increased to 0.417 nm. It suggested that the hyperbranched polytriazine were successfully grafted onto the surface of reduced graphene oxide sheets to exfoliate them, which agreed well with the XPS and TEM analysis results.

Fig. 2 XRD patterns of GO, GO-NH₂, RGO and HBPT-RGO.

TEM provided direct observation for the morphology of RGO and HBPT-RGO. Fig. 3a showed that RGO had a typical shape resembling the exfoliated crumpled thin flake, and the size of individual sheets could beyond many tens micrometers. However, the HBPT-RGO [Fig. 3b] showed a different morphology to RGO. After functionalized, the surface of HBPT-RGO appeared to be rough and clearly covered by the dark ball-like addends, which might be attributed to the grafting of hyperbranched polytriazine.^38,39 The AFM images of GO and HBPT-RGO could be seen in Fig. S3.†

Fig. 3 TEM images of RGO (a) and HBPT-RGO (b).

3.2. Mechanical properties of the composites

The dependency of the impact strength of the HBPT-RGO/BMI-BA composites on the content of the HBPT-RGO was shown in Fig. 4a. It could be seen that suitable amount of HBPT-RGO could properly improve the impact strength of BMI-BA resin. The impact strength of the composites increased continuously with the addition of HBPT-RGO, and reached the maximum value by 16.7 kJ m⁻² at 0.6 wt% HBPT-RGO. When the amount of HBPT-RGO was further increased, unfortunately, the impact strength of the composites decreased. This phenomenon could be explained that excessive HBPT-RGO could not be well dispersed in the BMI-BA matrix and agglomerated to cluster. The Fig. 4b showed the impact strength of the neat BMI-BA and graphene-based composites. Compared to the impact strength of neat BMI-BA, that of RGO/BMI-BA composites decreased slightly (detail data in Fig. S7†), the impact strength of HBPT-RGO/BMI-BA composite increased as much as 56.1% in comparison with neat BMI-BA resin (10.7 kJ m⁻²).

Fig. 4 Impact strength of HBPT-RGO/BMI-BA composites with different content of HBPT-RGO (a), impact strength of neat BMI-BA resin and BMI-BA composites with 0.6 wt% RGO as well as 0.6 wt% HBPT-RGO (b) and the flexural strength of HBPT-RGO/BMI-BA composites with different content of HBPT-RGO (c).

To further research the fracture characteristics of the BMI-BA composites, the fracture surfaces of neat BMI-BA resin, the composites filled with 0.6 wt% RGO and 0.6 wt% HBPT-RGO taking from impact tests were investigated using SEM (Fig. 5). The crack propagation regions of neat BMI-BA resin and its composites were different. The fracture surface of neat BMI-BA resin showed in Fig. 5a was comparatively slippery, which exhibited a typical brittle feature, while that of the HBPT-RGO/BMI-BA composite was indented and containing many ductile sunken areas, indicating a typical rough feature, which was considered from a large number of interfaces introduced by HBPT-RGO (Fig. 5c). As shown in Fig. 5b and c, a distinct difference in the interfacial interactions between the BMI-BA matrix and the nano-particles. The RGO sheets protruded cleanly from the fracture surface, and the micro-crack caused by pull-out of nanosheets indicating a weak interfacial bond. Compared with RGO, the protruding HBPT-RGO was thickly coated with absorbed polymer and intercalated into polymer matrix, indicating strong matrix–HBPT-RGO interactions which lead to better mechanical properties.

Fig. 5 SEM images of fracture surfaces taken from neat BMI-BA (a), the composites with 0.6 wt% RGO (b) and 0.6 wt% HBPT-RGO (c) in the impact tests.

It was suggested that two main differences between HBPT-RGO and RGO leaded to these results. Firstly, compared with the relatively inert surface chemistry of RGO, HBPT-RGO contained abundant amino groups which could form chemical bonds with the carbon–carbon double bonds of BMI-BA as a whole (the reaction mechanism was shown in Fig. 6). The corresponding mechanisms of the reactions had been discussed by many researches.^40,41 These chemical bonds contributed to the efficient load transfer between matrix and nanofillers, and absorbed large amounts of energy to prevent or bridge the growth of micro-crack in matrix effectively.¹⁴ Together with the high surface area of HBPT-RGO, these surface chemical bonds leaded to stronger interfacial interactions with BMI-BA and thus substantially larger influence on the properties of the neat resin. Secondly, the existence of abundant polar functional groups on the HBPT-RGO surface, as well as the extremely small thickness of the resulting HBPT-RGO lead to a homogenous dispersion of nanosheets in matrix.⁴² This state of dispersion probably resulted in an enhanced mechanical interlocking with the polymer chains and better adhesion. These strong interfacial interactions enabled graphene to bring its amazing mechanical properties into effect, achieving toughness of BMI-BA resin. Moreover, the chemical bonds between HBPT-RGO and matrix slightly reduced the crosslink density of BMI-BA resin, which also enabled the polymer chains to slip and provide more free volume under impact. The performance of RGO/BMI-BA composites degraded due to agglomeration of nanosheets and weak interfacial bond.⁴³

Fig. 6 Schematic of the reactions between HBPT-RGO and BMI-BA.

The flexural strength for the neat resin and composites was shown in Fig. 4c. By comparison, the composites containing the HBPT-RGO exhibited higher strength values than the RGO counterparts over the whole filler contents study. It could be seen that the small addition of HBPT-RGO increased the flexural strength of BMI-BA resin. The flexural strength increased continuously with the addition of HBPT-RGO, and reached a maximum value by 157.42 MPa at 0.4 wt% HBPT-RGO, which increased as much as 15.97% in comparison with that of neat BMI-BA (135.74 MPa). However, when the loading of flexural strength was further increased, the flexural strength decreased. And when the HBPT-RGO content was greater than 0.8 wt%, the flexural strength of composites would be lower than that of the neat BMI-BA composite. This phenomenon could be explained that excessive HBPT-RGO agglomerated in the matrix (see Fig. S9†), thus the advantages of the HBPT-RGO were not fully realized.⁴⁴ Meanwhile, there were more defects in HBPT-RGO because of the reduction and functionalization steps, which decreased the mechanical strength of graphene, so the strength and dispersion of nanofillers corporately leaded to the result that HBPT-RGO/BMI-BA composites presented this mechanical performance. The studies of Zhao et al. and Liang et al. also had the similar results when reached maximum impact strength at higher loading and flexural strength at lower loading.^17,45 To reveal the composition of the composite with 0.6 wt% HBPT-RGO, the reaction of composites was investigated by FTIR shown in Fig. 7. A series of constant absorption peaks observed at 3415, 1783, 1510 and 1378 cm⁻¹ were attributed to the –OH stretching vibration of BA, the antisymmetric stretching vibration of C=O, the N–H out-of-plane bending vibration and the O=C–N stretching vibration of BMI-BA. Almost every absorption peak did not change, which was mainly due to the very small amount of addition of the filler. However, it was noteworthy that absorptions at 1105 cm⁻¹ and 1258 cm⁻¹ were assigned to symmetric and anti-symmetric stretching vibration absorption of C–O–C increased with the addition of HBPT-RGO, which was attributed to that the C–O–C band in the BMI-BA system overlapped with the Si–O–C in the HBPT-RGO and the C–N band produced by the chemical reactions that occurred between HBPT-RGO and BMI-BA.⁴⁶ Owing to the reactions above, the adhesion between BMI-BA and HBPT-RGO could be further improved, thus the composites exhibited excellent mechanical properties. To further confirm this reaction occurred in the preparation of composites, we had investigated the composites by XRD (see Fig. S10†).

Fig. 7 FTIR spectra of the neat BMI-BA and HBPT-RGO/BMI-BA composite.

3.3. Tribological properties of the composites

Fig. 8a showed the friction coefficients of HBPT-RGO/BMI-BA composites as a function of HBPT-RGO content for steady-state sliding against the counterpart steel ring under dry conditions. It could be seen that, the friction coefficient of BMI-BA resin could be obviously decreased with the addition of HBPT-RGO during the sliding time, and the formation time of transform film on the surface of counterpart steel ring was also shortened. When the content of HBPT-RGO was 0.6 wt%, the composites exhibited the more stable frictional coefficient than other content of HBPT-RGO during the friction process, which indicated that reasonable addition of HBPT-RGO formed uniform transform film. However, while the HBPT-RGO content was above 0.6 wt%, the friction coefficient of that displayed a slight increase.

Fig. 8 The friction coefficient (a) and the volume wear rate (b) of the composites with different content of HBPT-RGO.

Fig. 8b indicated the effects of HBPT-RGO content on the volume wear rate of HBPT-RGO/BMI-BA composites. It could be clearly seen that the volume wear rate decreased continuously with the suitable addition of HBPT-RGO. When the content of HBPT-RGO was 0.6 wt%, the volume wear rate of the composite reached the lowest value by 2.7 × 10⁻⁶ mm³ N⁻¹ m⁻¹, which decreased as much as 84.1% compared to BMI-BA resin (15.9 × 10⁻⁶ mm³ N⁻¹ m⁻¹). This result was in agreement with the test results of impact strength, which mainly attributed to the HBPT-RGO/BMI-BA composites containing 0.6 wt% HBPT-RGO could absorb more energy during the wear process. However, when the content of HBPT-RGO was further increased, the volume wear rate of the composite could be attributed to the enhancement of its mechanical properties. With the improvement of the mechanical properties, the load-carrying capability of the composites increased, thus the deformation of the composites could be reduced effectively during the wear process.⁴⁷ Consequently the wear resistance of the composites with the addition of HBPT-RGO was enhanced.

3.4. Wear mechanism of materials

In order to further investigate the effects of HBPT-RGO on the wear and friction behavior of HBPT-RGO/BMI-BA composites, the morphologies of the worn surfaces were observed by SEM. Fig. 9a showed the worn surface of neat BMI-BA. It was rough and many scratching grooves were also found, which reflected that the composite had relatively poor wear resistance in its sliding against the steel counterpart and its wear mechanism mainly followed adhesive wear mechanism. The worn surface of 0.6 wt% RGO/BMI-BA composite could be seen in Fig. S11.† In Fig. 9b, fewer cracks and scales were found on the worn surface of 0.6 wt% HBPT-RGO/BMI-BA composite and it seemed more compact and less apt to form debris from the matrix, which might be related with the superior tribological property of graphene and better interface bonding strength between HBPT-RGO and BMI-BA resin mentioned above, which was the characteristic of abrasive wear mechanism. It also indicated that the anti-wear ability of the composites was reinforced by HBPT-RGO. There were also some HBPT-RGO particles that could be seen on the worn surface, which indicated that HBPT-RGO near the surface of resin matrix could be precipitated and act as a self-lubricating “carbon film” on the worn surface of composites during the wear process (see Fig. S13 and S14†). Meanwhile, the “carbon film” inhibited the transfer of resin to the worn surface of composites and the counterpart steel ring during the wear process, thereby improving the friction-reducing and anti-wear capacity of the resin composites.^48,49 In addition, the good dispersion and interface bonding strength of HBPT-RGO in the BMI-BA matrix possibly gave a uniform table and durable lubrication transfer film between the composites and the counterpart steel ring. The transfer film could protect the material during the sliding process, which resulted in lower frictional coefficient and higher wear resistance.

Fig. 9 SEM images of worn surfaces taken from neat BMI-BA (a) and the composites with 0.6 wt% HBPT-RGO (b) after friction test.

3.5. Thermal stability of the composites

TGA results could further confirm the above phenomena. As shown in Fig. 10, the decomposition temperature (T_d, at 5.0 wt% weight loss) of the composites was measured and compared. At fixed loading of 0.6 wt%, the composites filled with RGO and HBPT-RGO showing 389.5 and 409.4 °C increased in T_d values in comparison with that (372.8 °C) of the neat BMI-BA. It indicated that the hyperbranched polytriazine functionalized reduced graphene oxide was effective to enhance the thermal stability of bismaleimide resin. The increased thermal stability could be explained in terms of the dispersion of the reduced graphene oxide sheets and interfacial interaction with the bismaleimide matrix. The improved dispersion and interfacial interactions of HBPT-RGO in the BMA-BA matrix reduced the mobility of the local matrix around the reduced graphene oxide sheets, offering a better barrier effect of reduced graphene sheets to retard the volatilization of polymer decomposition products.⁵⁰ Moreover, the decreased residual weight of the composites containing the HBPT-RGO was observed at 800 °C compared with the corresponding composites containing the RGO, which might be due to the thermal degradation of grafted hyperbranched polytriazine on the reduced graphene oxide surface. The excellent thermal stability could protect the composites from the harm of high heat during the wear process, thus adhesive wear in turn could be inhibited effectively (see Fig. S15†).^51,52

Fig. 10 TGA curves of neat BMI-BA, the composites with 0.6 wt% RGO and 0.6 wt% HBPT-RGO.

4. Conclusions

A kind of high-performance composite had been fabricated in situ, the BMI-BA resin was chosen as the matrix and the reduced graphene oxide chosen as the nanofiller and polyamine curing agent. The reduced graphene oxide was modified with hyperbranched polytriazine before use in order to ensure the nanosheets could uniformly disperse in the polymer matrix and improve the interface interaction between the nanosheets and polymer matrix in the composites. The HBPT-RGO/BMI-BA composites displayed better mechanical and tribological properties than those of the composites with RGO, especially at the HBPT-RGO content of 0.6 wt%. Moreover, the HBPT-RGO/BMI-BA systems exhibited better thermal stability than that of neat BMI-BA resin. In light of its high tribological performance, the HBPT-RGO/BMI composites have great potential to be applied as wear-resistant composites in many fields.

Acknowledgements

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

Notes and references

1. K. Salini, M. Krishna, H. N. Narasimha Murthy, K. S. Rai Satheesh and M. Chandran, Polym.–Plast. Technol. Eng., 2010, 49, 1390.
2. X. W. Tian and Z. G. Yang, Polym.–Plast. Technol. Eng., 2011, 50, 474.
3. C. Liu, H. X. Yan, Z. Y. Chen, L. X. Yuan and T. Y. Liu, J. Mater. Chem. A, 2015, 3, 10559.
4. L. Yuan, A. Gu, G. Liang and Z. Zhang, Compos. Sci. Technol., 2008, 68, 2107.
5. X. Liu, Y. Yu and S. Li, Eur. Polym. J., 2006, 42, 835.
6. Y. Zhang, S. Qi and F. Zhang, J. Reinf. Plast. Compos., 2012, 31, 915.
7. H. Tang, X. Fan, Z. Shen and Q. Zhou, Polym. Eng. Sci., 2013, 54, 1675.
8. Y. F. Fu, K. Xu, Z. Y. Sun, F. Q. Zhang, D. M. Chen and J. Li, Polym.–Plast. Technol. Eng., 2012, 51, 1068.
9. Q. Cheng, J. Bao, J. Park, Z. Liang, C. Zhang and B. Wang, Adv. Funct. Mater., 2009, 19, 3219.
10. C. Liu, H. X. Yan, L. X. Yuan, Z. Y. Chen and M. M. Zhang, J. Polym. Sci., Part A: Polym. Chem., 2015 DOI:10.1002/pola.27661.
11. A. Vyas and J. O. Iroh, J. Appl. Polym. Sci., 2013, 130, 3319.
12. A. Vyas and J. O. Iroh, Polym. Compos., 2015 DOI:10.1002/pc.23399.
13. L. B. Zhang, J. Q. Wang, H. G. Wang, Y. Xu, Z. F. Wang, Z. P. Li and S. R. Yang, Composites, Part A, 2012, 43, 1537.
14. Y. J. Wan, L. X. Gong, L. C. Tang, L. B. Wu and J. X. Jiang, Composites, Part A, 2014, 64, 79.
15. M. Zhang, H. X. Yan, X. Yang and C. Liu, RSC Adv., 2014, 4, 45930.
16. H. Wang, B. Xia, Y. Yan, N. Li, J. Y. Wang and X. Wang, J. Phys. Chem. B, 2013, 117, 5606.
17. C. Y. Chang, S. P. Ju, J. W. Chang, S. C. Huang and H. W. Yang, RSC Adv., 2014, 4, 26074.
18. X. J. Shen, X. Q. Pei, S. Y. Fu and K. Friedrich, Polymer, 2013, 54, 1234.
19. X. J. Shen, X. Q. Pei, Y. Liu and S. Y. Fu, Composites, Part B, 2014, 57, 120.
20. Y. H. Zhao, Z. K. Wu and S. L. Bai, Composites, Part A, 2015, 72, 200.
21. B. L. Pan, G. Q. Xu, B. W. Zhang, X. J. Ma, H. Z. Li and Y. Z. Zhang, Polym.–Plast. Technol. Eng., 2012, 51, 1163.
22. G. Ren, Z. Zhang, X. Zhu, B. Ge, F. Guo, X. Men and W. Liu, Composites, Part A, 2013, 49, 157.
23. T. Ramanathan, A. A. Abdala, S. Stankovich, D. A. Dikin, M. Herrera-Alonso, R. D. Piner and L. C. Brinson, Nat. Nanotechnol., 2008, 3, 327.
24. Y. Tan, L. Fang, J. Xiao, Y. Song and Q. Zheng, Polym. Chem., 2013, 4, 2939.
25. Y. Huang, Y. Qin, N. Wang, Y. Zhou, H. Niu, J. Y. Dong and Y. Wang, Macromol. Chem. Phys., 2012, 213, 720.
26. W. H. Liao, S. Y. Yang, J. Y. Wang, H. W. Tien, S. T. Hsiao, Y. S. Wang and Y. F. Wu, ACS Appl. Mater. Interfaces, 2013, 5, 869.
27. S. Bai, X. Shen, G. Zhu, Z. Xu and Y. Liu, Carbon, 2011, 49, 4563.
28. Y. Zhan, X. Yang, H. Guo, J. Yang, F. Meng and X. Li, J. Mater. Chem., 2012, 22, 5602.
29. H. X. Yan, S. Li, Y. Jia and X. Y. Ma, RSC Adv., 2015, 5, 12578.
30. C. Wu, X. Huang, G. Wang, L. Lv, G. Chen, G. Li and P. Jiang, Adv. Funct. Mater., 2013, 23, 506.
31. C. Liu, H. Yan, S. Feng, T. Li and M. Zhang, Chem. Lett., 2014, 43, 1263.
32. N. I. Kovtyukhova, P. J. Ollivier, B. R. Martin, T. E. Mallouk, S. A. Chizhik, E. V. Buzaneva and A. D. Gorchinskiy, Chem. Mater., 1999, 11, 771.
33. J. Shen, B. Yan, T. Li, Y. Long, N. Li and M. Ye, Composites, Part A, 2012, 43, 1476.
34. B. Yuan, C. Bao, L. Song, N. Hong, K. M. Liew and Y. Hu, Chem. Eng. J., 2014, 237, 411.
35. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. V. Dubonos and A. A. Firsov, Science, 2004, 306, 666.
36. B. Ahmadi-Moghadam, M. Sharafimasooleh, S. Shadlou and F. Taheri, Mater. Des., 2015, 66, 142.
37. T. K. Gupta, B. P. Singh, R. K. Tripathi, S. R. Dhakate, V. N. Singh, O. S. Panwar and R. B. Mathur, RSC Adv., 2015, 5, 16921.
38. K. J. Peng, K. H. Wang, K. Y. Hsu and Y. L. Liu, J. Polym. Sci., Part A: Polym. Chem., 2014, 52, 1588.
39. S. H. Che Man, S. C. Thickett, M. R. Whittake and P. B. Zetterlund, J. Polym. Sci., Part A: Polym. Chem., 2013, 51, 47.
40. N. Regnier, M. Fayos and E. Lafontaine, J. Appl. Polym. Sci., 2000, 78, 2379.
41. A. Selvaganapathi, M. Alagar and P. Gnanasundaram, High Perform. Polym., 2013, 25, 622.
42. N. Hong, J. Zhan, X. Wang, A. A. Stec, T. Richard Hull, H. Ge and Y. Hu, Composites, Part A, 2014, 64, 203.
43. Y. Hwang, M. Kim and J. Kim, Composites, Part A, 2013, 55, 195.
44. T. K. Bindu Sharmila, N. B. Nair, B. T. Abraham, P. M. Beegum and E. T. Thachil, Polymer, 2014, 55, 3614.
45. K. Guo, Z. Hu, H. Song, X. Du, L. Zhong and X. Chen, RSC Adv., 2015, 5, 5197.
46. Y. Jia, H. X. Yan, L. Ma and J. P. Zhang, J. Polym. Res., 2014, 21, 1.
47. Y. Mo, M. Yang, Z. Lu and F. Huang, Composites, Part A, 2013, 54, 153.
48. Q. Lin, L. Qu, Q. Lü and C. Fang, Polym. Test., 2013, 32, 330.
49. M. Liu, Y. Duan, Y. Wang and Y. Zhao, Mater. Des., 2014, 53, 466.
50. R. Qian, J. Yu, C. Wu, X. Zhai and P. Jiang, RSC Adv., 2013, 3, 17373.
51. M. Zhou, H. Bi, T. Lin, X. Lü, F. Huang and J. Lin, J. Mater. Chem. A, 2014, 2, 2187.
52. X. Wang, S. Zhou, W. Xing, B. Yu, X. Feng, L. Song and Y. Hu, J. Mater. Chem. A., 2013, 1, 4383.

---

Footnote

† Electronic supplementary information (ESI) available: Figures about the morphologies, structures and properties of different composites. See DOI: 10.1039/c5ra06009e

---