Flexural properties, interlaminar shear strength and morphology of phenolic matrix composites reinforced with xGnP-coated carbon fibers

Abstract

In the present study, exfoliated graphite nanoplatelets (xGnP) with different particle sizes were coated onto polyacrylonitrile-based carbon fibers by a direct coating method. The flexural properties, interlaminar shear strength, and the morphology of the xGnP-coated carbon fiber/phenolic matrix composites were investigated in terms of their longitudinal flexural strength and modulus, interlaminar shear strength, and by optical and scanning electron microscopic observations. The results were compared with a phenolic matrix composite counterpart prepared without xGnP. The flexural properties and interlaminar shear strength of the xGnP-coated carbon fiber/phenolic matrix composites were found to be higher than those of the uncoated composite. The flexural and interlaminar shear strengths were affected by the particle size of the xGnP, while the particle size had no significant effect on the flexural modulus. It seems that the interfacial contacts between the xGnP-coated carbon fibers and the phenolic matrix play a role in enhancing the flexural strength as well as the interlaminar shear strength of the composites.

1. Introduction

Carbon materials with high surface areas such as carbon nanofibers, exfoliated graphite nanoplatelets (xGnP), carbon nanotubes (CNT), and graphene are representative of the carbon materials that can be widely used as conducting fillers [1-3]. The unique electrical and structural properties of these carbon nanoparticles have made them very attractive in automotive, aerospace, and electronic applications. Carbon nanoparticle-filled polymer composites have advantages over traditional polymer composites in their mechanical and thermal properties. It has been found that xGnP-filled polymer composites exhibit excellent electrical and thermal conductivities as well as good mechanical properties [4,5] even though they had a low xGnP content of less than 2 vol%. This effect is attributed to the extremely large surface area and the aspect ratio of the xGnP, which leads to the formation of a percolated conducting network within the polymer matrix at concentrations of less than 2 vol% [6,7].

The xGnP can be produced by the expansion and exfoliation of graphite flakes intercalated with highly concentrated acids, and they can be expanded up to a few hundred times their initial volume by thermal treatment or microwave processing. After the appropriate exfoliation process, these particles normally consist of multiple layers of graphene sheets. The xGnP exhibits very high modulus and stiffness along its graphene basal plane, as well as excellent electrical and thermal conductivities, due to its layered structure. Consequently, the xGnP can provide significant advantages in the development of conducting polymers or composites with effective electromagnetic interference shielding and conductivity, as well as enhancing the mechanical and thermal properties of relevant plastics and composites.

Li et al. [8] reported the effect of exfoliation and ultraviolet (UV)/ozone treatment of graphite on the conductivity of xGnP/epoxy composites, and the effect of hybrid CNT and xGnP particles on the mechanical and electrical properties of epoxy nanocomposites [9]. Kalaitzidou et al. [10] stressed the potential use of xGnP as a reinforcement to produce multifunctional polymer composites. They found that the large aspect ratio of xGnP increases the oxygen barrier property of polypropylene composites and significantly enhances the thermal conductivity of the polymer matrix. Fukushima explored the mechanical properties of chemical group-grafted xGnP incorporated epoxy composites [11]. It has been reported that acrylamide grafting treatment enhances the dispersion and adhesion of xGnP in epoxy matrices, and the resulting composites showed better mechanical properties than those produced with commercially available carbon fibers.

The use of polyacrylonitrile-based carbon fiber/phenolic matrix composites in thermal insulation applications have often been limited because of their low mechanical and interfacial properties. In order to improve the mechanical and interfacial properties of carbon fiber/phenolic matrix composites, many studies have attempted to enhance the interfacial adhesion between the reinforcing fiber and the matrix by using various fibertype fillers treated with sizing agents, or by modifying the fiber surface using chemical treatments [12-17]. For thermal insulation applications it has frequently been useful to produce phenolic matrix composites reinforced with xGnP-coated carbon fibers which have robust mechanical and electrical properties. One possible approach to obtain such a phenolic matrix composite is to anchor the xGnP particles on the carbon fiber surfaces. A variety of methods to anchor the particles on the carbon fiber surfaces have been used, such as direct coating, chemical vapor deposition and electrophoretic deposition [18,19].

Consequently, the objective of the present study is to investigate the flexural and interlaminar shear properties of phenolic matrix composites reinforced with polyacrylonitrile-based carbon fibers coated with xGnP of different particle sizes by a direct coating method, and to compare them to the properties of uncoated carbon fiber/phenolic counterparts.

 

2. Experimental

2.1. Materials

The xGnP used in this work was supplied from the Composites Materials and Structures Center of Michigan State University, East Lansing, MI, USA. Grafguard 160-50A (GrafTech, Independence, OH, USA) was used as a starting material for preparing the xGnP. It is graphite intercalated with a mixture of sulfuric acid and nitric acid at about 20 wt%. For the exfoliation process the intercalated graphite was treated for about 5 min in a conventional microwave oven. The microwave-treated graphite nanoplatelets were added to isopropanol and then ultrasonicated for 2 h, followed by a pulverization process to reduce the particle size. After filtering to separate the xGnP, the obtained material was dried in a convection oven. To prepare xGnP particles of less than 1 μm in average size, an additional milling process was performed with the xGnP for 4 d, using a vibratory ball-milling machine. To examine the effect of the aspect ratio of graphite nanoplatelets on the properties of xGnP-coated carbon fiber/phenolic composites, two types of xGnP were used: xGnP with a thickness of 10 nm and a diameter of 15 μm, and xGnP with the same thickness and an average diameter of 1 μm. Details of the xGnP exfoliation process are described elsewhere [11]. A resole-type phenolic resin SC-1008 (Borden Co., USA) was used as the matrix. The solid contents are 59% and the viscosity is 350 cps at 25℃. The gel time is 178 s and the maximum cure temperature is 165℃.

2.2. Preparation of xGnP-coated carbon fibers and carbon fiber/phenolic matrix composites

A coating tower machine was used to evenly coat the xGnP particles on the carbon fiber surfaces by a direct coating method. Unsized carbon fibers (12k, AS4; Hercules, Wilmington, DE, USA) were used in this study. Epoxy resin (diglycidyl ether of bisphenol A, DGEBA) (Epon 828, Shell Chemical, Houston, TX, USA) of about 0.5 wt% was used as a sizing agent for treating ‘as-received’ or uncoated, and xGnP-coated, carbon fibers. For the xGnP-coated carbon fibers, one important process step is to make the suspension solution homogeneous. About 0.5 wt% epoxy resin and 1 wt% xGnP were well mixed in isopropanol. The mixture was uniformly dispersed for 1 h by using an ultrasonicating apparatus. The heating temperature and the coating speed of the coating tower machine were 250℃-300℃ and 1.4 m/min, respectively.

Unidirectional (UD) carbon fiber/phenolic resin prepregs were made by using a tailor-made laboratory-scale prepregging machine. An important step in making the prepregs is to successfully impregnate the resole-type phenolic resin of above 40% into the carbon fibers by controlling the die gap. The die gap was 0.15 mm. The resin contents of carbon/phenolic prepregs were in the range of 31-38 wt%. The flexprepregs were dried at 100℃ for 30 min in a drying oven to volatilize the solvent therein.

UD-type xGnP-coated carbon fiber/phenolic matrix composites were fabricated at 160℃ for 2 h using an Autopress (Tetrahedron MTP-14, Tetrahedron Associates, San Diego, CA, USA). The curing profile of carbon fiber/phenolic matrix composites is shown in Fig. 1. The debulking process was carried out at 100℃ for 30 min to remove possible entrapped air and voids in the resulting composites.

Fig. 1.A time-temperature-pressure profile for curing exfoliated graphite nanoplatelets (xGnP)-coated carbon fiber/phenolic matrix composites.

2.3. Characterization

The density of carbon fiber/phenolic matrix composites was measured, based on the Archimedes principle according to American Society for Testing and Materials (ASTM) D792-86. The liquid chosen was deionized water. Three-point flexural tests were performed by using a universal testing machine (UTM, SFM-14, United Testing System Inc., California, USA) at the crosshead speed of 1.5 mm/min according to ASTM D-790. The specimen dimensions for the flexural test were 50.8 mm × 13 mm × 2.3 mm and the span-to-depth ratio was 16. In order to explore the interlaminar shear strength (ILSS) of the carbon fiber/phenolic matrix composites, short-beam shear tests were carried out according to ASTM D-2344. The crosshead speed was 1.3 mm/min and the span-to-thickness ratio was 4. Average values of the flexural properties and the ILSS were obtained from ten specimens, respectively.

The microstructure of the specimens obtained after the mechanical testing was observed by using an environmental scanning electron microscope (ESEM; Philips Electroscan 2020, Wilmington, MA, USA) to examine the longitudinal and fractured fiber surfaces with uncoated and coated carbon fiber/phenolic composites with different particles sizes of xGnP. The specimens were gold-coated prior to the microscopic observation by a sputtering method. Also, an optical microscope was used to observe the cross-section of the composites.

 

3. Results and Discussion

3.1. Characteristics of xGnP-coated carbon fibers and the composites

The xGnP-coated carbon fibers were prepared by passing the xGnP/epoxy/isopropanol suspension through a coating tower machine. Fig. 2 shows the ESEM micrographs of the uncoated and xGnP-coated carbon fibers. The epoxy resin of 0.5 wt% and the xGnP of 1 wt% were coated onto the surfaces of the carbon fibers.

Fig. 2.ESEM micrographs of ‘as-received’ carbon fibers (a) and exfoliated graphite nanoplatelets (xGnP)-coated carbon fibers; (b) xGnP of 1 μm; (c) xGnP of 15 μm in average size. ESEM, environmental scanning electron microscope.

The physical characteristics of the xGnP-coated carbon fiber/phenolic matrix composites are summarized in Table 1. The phenolic resin content of the 1 wt% xGnP-coated carbon fiber/phenolic matrix composites is about 23-24 wt%. The thickness is about 2.18-2.31 mm. The composite density is about 1.48-1.54 g/cm3. It seems that for the composite with 1 μm sized xGnP the resin content and the thickness are lower and the density is higher than the cases without xGnP, and with 15 μm sized xGnP.

Table 1.xGnP, exfoliated graphite nanoplatelets.

3.2. Mechanical properties of xGnP-coated carbon fiber/phenolic matrix composites

Fig. 3 displays the longitudinal flexural properties of the phenolic matrix composites reinforced with carbon fibers coated with xGnP of different particle sizes. The longitudinal flexural strength of the xGnP-coated carbon fiber/phenolic matrix composites is about 18%-40% higher than that of the uncoated composite. This indicates that the xGnP plays a significant role in reinforcing the interfacial region between the carbon fibers and the phenolic matrix, and contributes to healing some cracks. As the xGnP particle size increases from 1 μm to 15 μm, the longitudinal flexural strength of the xGnP-coated carbon fiber/phenolic matrix composites is slightly increased. The longitudinal flexprepregs ural modulus of the composite with carbon fibers coated with xGnP is higher than that of the uncoated counterpart. However, the particle size effect of xGnP is not significant for the flexural modulus. This may be explained in that for both cases with different particle sizes, the xGnP with the high length/thickness ratio positively contributed to increasing the composite modulus, which is mainly influenced by the alignment of the layered structure.

Fig. 3.The variation in the longitudinal flexural strength and modulus of exfoliated graphite nanoplatelets (xGnP)-coated carbon fiber/phenolic matrix composites with xGnP of different particle sizes.

Fig. 4 shows the variation in the ILSS value measured for the carbon fiber/phenolic matrix composites with xGnP of different particle sizes. With increasing xGnP particle size, the ILSS of the xGnP-coated carbon/phenolic matrix composites is about 10%-13% higher than that of the uncoated composite. It is obvious that the xGnP-coated carbon fibers help increase the interfacial strength of the carbon fiber/phenolic matrix composites. It may be that anchoring the xGnP particles to the carbon fiber surfaces with the assistance of an epoxy sizing agent, as illustrated in Fig. 5a, plays a positive role not only in reinforcing the composite but also in improving the interfacial adhesion between the carbon fibers and the phenolic matrix.

Fig. 4.The variation in the interlaminar shear strength of exfoliated graphite nanoplatelets (xGnP)-coated carbon fiber/phenolic matrix composites with xGnP of different particle sizes.

Fig. 5.Schematic illustrating the carbon fibers coated with the exfoliated graphite nanoplatelets (xGnP) anchored by epoxy sizing (a) and the interfacial region with the xGnP between the fibers and the matrix (b).

On the other hand, when the xGnP particle size was increased from 1 μm to 15 μm the ILSS was slightly decreased. This may be due to the xGnPs with relatively large particle size, which may more or less prevent efficient interfacial bonding between the fibers and the matrix, based on the schematic mechanism proposed by Drzal et al. [20], as indicated in Fig. 5b. However, the ILSS of the xGnP-coated composites is higher than that of the uncoated ones because the interfacial adhesion was increased by the xGnP coating layer on the carbon fiber surfaces.

The optical microscopic observations of the cross-section of the uncoated and xGnP-coated carbon fiber/phenolic matrix composites are presented in Fig. 6. A number of voids can be observed in both the uncoated and xGnP-coated composites. This is because volatiles are generated during the curing process and entrapped in the resulting composite.

Fig. 6.Optical micrographs of the cross-section of ‘as-received’ (a) and exfoliated graphite nanoplatelets (xGnP)-coated carbon fiber/phenolic matrix composites: (b) 1 μm sized xGnP; (c) 15 μm sized xGnP.

No microscopic evidence can be found in the cross-section images showing any effect of the xGnP between the carbon fibers and the phenolic matrix for the uncoated and xGnP-coated composites.

The ESEM observation of the fractured surfaces of each composite after the flexural testing illustrates the effect of the xGnP coating in the composites. Fig. 7 shows the ESEM micrographs of the fractured cross-sections of the uncoated and xGnP-coated carbon fiber/phenolic matrix composites. It is likely that all of the specimens exhibit a resin-starved matrix region due to low resin content. Consequently, the interfacial region between the fibers and the matrix is not well developed for the sake of the clean surfaces of the fractured carbon fibers. Nevertheless, the flexural and interlaminar shear properties were increased to some extent in the composites with xGnP.

Fig. 7.ESEM micrographs of the fractured fiber cross-section of ‘asreceived’ (a) and exfoliated graphite nanoplatelets (xGnP)-coated carbon fiber/phenolic matrix composites after the flexural testing; (b) 1 μm sized xGnP; (c) 15 μm sized xGnP. ESEM, Environmental scanning electron microscope.

The anchoring of xGnP onto the carbon fiber surfaces by epoxy sizing may positively contribute to the improving the above-mentioned properties. The result implies that some marginal increase in the interfacial strength between the carbon fiber and the phenolic matrix may be attainable. It is suggested that incorporating chemically functionalized graphite or graphene nanoplatelets [21] into the carbon fiber surfaces may be beneficial for forming additional interfacial bonding between the fibers and the matrix in the composite, and also for improving the mechanical properties of the resulting carbon fiber/phenolic matrix composite.

 

4. Conclusions

xGnP-coated carbon fibers were prepared by a direct coating method, and the flexural properties, ILSS and the morphology of the phenolic matrix composites reinforced with carbon fibers coated with xGnP of different particle sizes were investigated. The longitudinal flexural strength of the xGnP-coated carbon fiber/phenolic matrix composites was about 18%-40% higher than that of the uncoated composite counterpart. The ILSS value of the xGnP-coated carbon fiber/phenolic matrix composites was about 10%-13% higher than that of the uncoated one. The flexural strength was higher with the large particle size xGnP, whereas the ILSS was slighly higher with the small particle size xGnP.