Chemical Physics Letters 703 (2018) 8–16
Contents lists available at ScienceDirect
Chemical Physics Letters journal homepage: www.elsevier.com/locate/cplett
Research paper
Study on interfacial and mechanical improvement of carbon fiber/epoxy composites by depositing multi-walled carbon nanotubes on fibers Chufan Xiao, Yefa Tan, Xiaolong Wang ⇑, Li Gao, Lulu Wang, Zehao Qi College of Field Engineering, Army Engineering University of PLA, Nanjing 210007, China
a r t i c l e
i n f o
Article history: Received 27 February 2018 In final form 8 May 2018 Available online 8 May 2018 Keywords: Epoxy resin Carbon fiber Carbon nanotubes Composite
a b s t r a c t To improve the interfacial properties between carbon fiber (CF) and epoxy resin (EP), T300 carbon fibers were coated with multi-walled carbon nanotubes (MWCNTs) using aqueous suspension deposition method. The carbon fiber/epoxy laminated composites were prepared by molding process. The wettability and interfacial properties between MWCNTs deposited carbon fibers (MWCNTs-T300) and EP were studied. The mechanical properties of carbon fiber/epoxy laminated composites were tested, and the mechanism of the interface strengthening was discussed. The results show that the surface energy of T300 carbon fiber is obviously increased after MWCNT deposition. The contact angle between MWCNTs-T300 and EP is reduced, and the interfacial energy and adhesion work are greatly improved. The MWCNTs-T300/EP laminated composites have excellent mechanical properties, the flexural strength is 822 MPa, the tensile strength is 841 MPa, and the interlaminar shear strength (ILSS) is 25.68 MPa, which are increased by 15.1%, 17.6% and 12.6% compared with those of the original carbon fiber/EP laminated composites (original T300/EP) respectively. The MWCNTs-T300/EP composites have good interface bonding performance, low porosity and uniform fiber distribution. Interfacial friction and resin toughening are the main mechanisms for the interface enhancement of MWCNTs-T300/EP composites. Ó 2018 Elsevier B.V. All rights reserved.
1. Introduction Carbon fiber (CF) reinforced epoxy composites have attracted wide attention in academia and engineering fields for their good mechanical and economic performance [1]. However, under the harsh working conditions of high load and strong impact, the carbon fiber reinforced epoxy composites still have many disadvantages, such as low strength, low toughness and high environmental sensitivity. The mechanical properties of carbon fiber reinforced epoxy composites are not only affected by the inherent characteristics of CF and epoxy resin (EP), but also affected by the interfacial interaction of CF and EP [2]. The interface between CF and EP is the key element in the performance of the composite [3]. In recent years, to improve the mechanical properties of composites, carbon nanotubes (CNTs) have been introduced into the traditional continuous fiber reinforced polymer composites to build hierarchical reinforcing structure, which is currently an important research topic [4]. Because of their good mechanical ⇑ Corresponding author. E-mail address:
[email protected] (X. Wang). https://doi.org/10.1016/j.cplett.2018.05.012 0009-2614/Ó 2018 Elsevier B.V. All rights reserved.
properties, CNTs are widely used as additives to improve the mechanical properties of carbon fiber reinforced plastics (CFRP) [5,6]. Researches have shown significant improvements in glass fiber and carbon fiber composites by using multi-scale reinforcements, especially the fiber-matrix interfacial properties such as interfacial shear strength (IFSS) [2,7–10], interlaminar fracture toughness [11,12] and glass transition temperature (Tg) [13]. Sager et al. reported that the interfacial shear strength of the radially aligned MWCNTs deposited carbon fiber specimens was increased by 11% than that of the untreated, non-sizing carbon fibers [8]. CNTs can effectively enhance the interaction of the CF-EP interface by mechanical interlocking of the EP matrix and CF surface [14]. In general, the mechanical properties of CFRP were improved after the addition of CNTs, due to the strengthening of the fiber-matrix interface [15]. Iwahori et al. studied the mechanical properties of CNT modified CF/EP laminated composites, and the flexural strength of CF/EP composite increased by 2.7% due to the addition of CNTs [16]. The reason for the low increase in flexural strength is that the CNTs were directly dispersed in the epoxy matrix rather than connected to the carbon fiber surface. Dispersing CNT into the composite matrix and directly connecting CNT to the fiber surface are two strategies for forming
C. Xiao et al. / Chemical Physics Letters 703 (2018) 8–16
CNT-based hierarchical composite. To further improve the interfacial properties between fiber and resin, many researchers have focused on attaching CNTs directly to fiber bundles and fabrics. So far, there have been four methods for directly attaching CNTs to fiber surface: (1) chemical reaction between functionalized CNTs and fiber [17,18], (2) electrophoretic deposition of CNTs on the fiber surface, (3) the growth of CNTs on the fiber by chemical vapor deposition (CVD), (4) depositing CNTs comprising coating the fibers [19] with a CNT-containing sizing (a thin polymer coating) and spraying the CNT-containing solution onto the fibrous fabric. Among the above methods, the chemical reaction method and electrophoretic deposition method all require complex chemical treatment and complicate equipment. The CVD method not only requires expensive equipment and complex processing technology, but also the high processing temperature, which will lead to surface damage of carbon fiber, and affecting the mechanical properties of carbon fiber. For example, the tensile test performed by Sager et al. showed that CVD treatment significantly reduced the tensile properties of carbon fibers, and the tensile strength and tensile modulus were reduced by 30% and 13%, respectively, in the case of random orientation of MWCNTs [8]. For the depositing CNTs method, it is the most simple and effective, and will not cause damage to the fibers (even repairing surface defects and improving fiber tensile properties) [19–21]. In addition, the process does not require the removal of industrial sizing and can be used for various types of carbon fiber reinforcements, such as fiber bundles, fabrics and preforms. To improve the interface performance of carbon fiber composite, some researches have adopted the method of depositing CNTs. For example, Kwon and colleagues achieved an increase in interlaminar fracture toughness of CF/vinyl ester laminate composites by depositing CNTs, with a maximum increase of about 30%. However, the current researches on depositing CNTs method is mainly focused on improving one single mechanical properties of composites, which cannot achieve the improvement of comprehensive mechanical properties, and the study on its strengthening mechanism is still at the level of qualitative analysis. There is a lack of quantitative analysis of the interfacial properties between carbon fiber and epoxy resin, and the influence of interfacial properties on the mechanical properties of carbon fiber/epoxy composites. To improve the interfacial properties of carbon fiber and epoxy resin, and further improve the mechanical properties of carbon fiber/epoxy composite, carbon fibers (T300) were coated with MWCNT using aqueous suspension deposition method. Carbon fiber/epoxy laminated composites were prepared by molding process. The surface morphology of MWCNT deposited carbon fiber (MWCNTs-T300) and the wettability and interfacial properties between MWCNTs-T300 and EP were studied. The mechanical properties of carbon fiber/epoxy laminated composites were tested. The tensile and bending fracture morphology were analyzed. The mechanisms of interface strengthening are discussed, which provides the theoretical basis and technical support for the preparation of carbon fiber/epoxy composites with excellent performance and their application in engineering field.
9
polyether amine (PEA), provided by Suzhou Chanco Industrial Co., Ltd. 2.2. Preparation of samples MWCNTs were dispersed in deionized water by ultrasonic (ultrasonic generator, sonic power 125 W, ultrasonic frequency 80 kHz) for 3 h prior to the preparation of MWCNT deposited CF to obtain a stable suspension at a concentration of 0.1 wt%. Then, the carbon fiber fabric was immersed in the suspension for 20 min, followed by drying at 70 °C for 2 h and drying at 120 °C for 2 h. The original CF laminated composite samples and MWCNT deposited CF laminated composite samples were prepared by compression molding using E-44 resin as the matrix and T300 carbon fiber as the reinforcing material. The size of the tensile samples and bent samples are 230 25 2 mm and 110 10 4 mm, respectively. The thickness and fiber content of the two kinds of samples are the same. The fiber volume fraction of the tensile samples is about 50%, and the fraction of the bent samples is about 60%. The main steps of the preparation are: tailoring the carbon fiber fabric, cleaning the mold, coating the mold with release agent and laying a layer of mold release paper. Then, the resin impregnated carbon fiber fabrics are stacked in the mold, close mold and pressurize, curing at 70 °C for 24 h. The samples are obtained after mold unloading. 2.3. Characterization After the MWCNT deposition, the CF fabric was torn to expose the inside of the fabric, and then the carbon fiber bundle was examined by the Hitachi S4800 scanning electron microscope (SEM) to evaluate the adhesion state of MWCNT on the carbon fiber surface. The tensile strength and modulus of the composites were measured by the electronic universal testing machine (CMT 5105) using rectangular specimens made according to the American Society for Testing and Materials (ASTM) standard D3039. Three-point bending test was carried out on the bend specimens by CMT 5105 electronic universal testing machine. The test standard is ASTM D790-2015 and the bending speed is 2 mm/min. In addition, interlaminar shear strength (ILSS) was tested according to ASTM standard D2344. The test specimens were placed in a plastic bag and pour into distilled water, seal and hold at 50 °C for 24 h. The test specimens were dried at 50 °C and then test the mechanical properties of the composite (test standard is the same as room temperature mechanical properties). At least 5 specimens were tested for each set of conditions. The heat resistance was measured by DSC (Mettler-Toledo, DSC823e) (N2 atmosphere, the test temperature is from room temperature to 250 °C, scanning rate of 10 °C/min, mass of 20 mg). The micromorphology of the sample fracture surface was detected by the Hitachi S4800 scanning electron microscope. 3. Results and discussion
2. Experimental 3.1. Surface characteristics of MWCNTs deposited carbon fiber 2.1. Materials T300 carbon fiber monodirectional fabric is provided by Nanjing Fiberglass Research & Design Institute Co., Ltd. MWCNT: S-MWNT1020, length 0.5–2 lm, diameter 10–20 nm, and purity greater than 97%, produced by Shenzhen Nanotech Port Co., Ltd. The E44 epoxy resin with an epoxy value of 0.44 mol/100 g is provided by Taizhou Huili Electronic Material Co., Ltd. Curing agent: D-230
The deposition state of MWCNTs on carbon fiber surface after deposition process was observed by SEM. Fig. 1a is the surface morphology of the original carbon fiber, indicating that the surface of the original carbon fiber is smooth. The typical SEM images of carbon fibers after MWCNT deposition are shown in Fig. 1b–d. In Fig. 1b, due to the low concentration of MWCNTs suspension (0.05 wt%), the MWCNTs deposited to the carbon fiber surface is
10
C. Xiao et al. / Chemical Physics Letters 703 (2018) 8–16
Fig. 1. SEM image: (a) original carbon fiber, (b) 0.05 wt% MWCNTs deposited carbon fiber, (c) 0.1 wt% MWCNTs deposited carbon fiber, (d) 0.15 wt% MWCNTs deposited carbon fiber.
less, which cannot completely cover the carbon fiber surface and are unevenly distributed. However, if the concentration of MWCNTs is too high (0.15 wt%), it’s unable to form a stable suspension, although the number of MWCNTs adhered on the carbon fiber surface increases, the distribution of MWCNTs is not uniform and serious agglomeration phenomenon occurs (Fig. 1d). When the concentration of MWCNTs in the suspension is appropriate (0.1 wt %), the MWCNTs are randomly oriented and uniformly distributed on the carbon fiber surface, and the attached MWCNTs provide a good coverage on the carbon fiber surface, as shown in Fig. 1c. Therefore, the concentration of MWCNTs suspension has a great influence on the deposition amount and state of MWCNTs on carbon fiber surface. It is necessary to properly control the concentration of MWCNTs suspension to deposit an enough uniformly distributed MWCNTs on carbon fiber surface, which will benefit to improve the interfacial properties of carbon fiber laminated composites. 3.2. Wettability and interfacial properties between carbon fibers and EP In order to further analyze the effect of MWCNT deposition on the wettability and interfacial properties between carbon fiber and EP, the contact angle test of different types of carbon fibers were conducted. The contact angle images of different carbon fibers are shown in Fig. 2. The contact angle between the carbon fiber and EP is shown in Table 1. The contact angle between the original T300 and EP is 47.1°. After the MWCNT deposition, the contact angle is reduced in great margin. For example, when the T300 carbon fibers are treated by 0.05 wt%, 0.1 wt% and 0.15 wt% MWCNT deposition, the contact angles between T300 and EP are decreased to 43.1°, 40.3° and 42.4°, respectively. Wherein the contact angle of the T300 carbon fiber deposited with 0.1 wt% MWCNTs is minimized (40.3°). This indicates that the wettability between carbon fiber
and EP has been improved effectively after MWCNT deposition, which can effectively reduce the interface defects caused by poor infiltration, such as void, so that the interface bonding performance between carbon fiber and EP is improved. Thus, the mechanical properties of carbon fiber/EP composites will be improved. Interfacial energy and adhesion work are important indexes to measure the interfacial properties of fiber and resin. To quantitatively analyze the interfacial properties of MWCNT deposited carbon fiber, the interfacial energy and adhesion work are calculated. Based on the Young-Dupre equation [22], the relationship between surface energy, contact angle (h) and adhesion work can be expressed as follows:
cLS ¼ cS cL cosðhÞ
ð1Þ
W a ¼ cL þ cS cLS ¼ cL ð1 þ coshÞ
ð2Þ
where cS, cL and cLS are the surface energy of solid, liquid and solidliquid interfaces, respectively. Wa is the adhesion work of solidliquid interface. According to the surface energy component method of Owens and Wendt [23], the solid-liquid interfacial energy can be expressed as:
qffiffiffiffiffiffiffiffiffiffi
qffiffiffiffiffiffiffiffiffiffi
cLS ¼ cS þ cL 2 cpS cpL 2 cdS cdL
ð3Þ
where cpS and cpL are the polar components of solid and liquid surface energy, respectively. cdS and cdL are nonpolar components. So
cS ¼ cpS þ cdS
ð4Þ
cL ¼ cpL þ cdL
ð5Þ
Base on Eqs. (1) and (3), Eq. (6) can be obtained:
qffiffiffiffiffiffiffiffiffiffi
qffiffiffiffiffiffiffiffiffiffi
cL ðcosh þ 1Þ ¼ 2 cpS cpL þ 2 cdS cdL
ð6Þ
11
C. Xiao et al. / Chemical Physics Letters 703 (2018) 8–16
Fig. 2. Contact angle images of different fibers: (a) original T300, (b) 0.05 wt% MWCNTs deposited T300, (c) 0.1 wt% MWCNTs deposited T300, (d) 0.15 wt% MWCNTs deposited T300.
Table 1 Contact angle between MWCNT deposited carbon fiber and epoxy resin. Fiber type
Original T300
0.05 wt% MWCNTs deposited T300
0.1 wt% MWCNTs deposited T300
0.15 wt% MWCNTs deposited T300
Contact angle (°)
47.1 ± 0.9
43.1 ± 2.3
40.3 ± 1.7
42.4 ± 1.6
According to Eq. (6), two kinds of solvents were used to determine the surface energy of carbon fiber. The solvents used in this experiment are glycerol and water, in which the cpL and cdL of glycerol are 21.7 and 42 mJ/m2 respectively, and the cpL and cdL of water are 51.7 and 21.1 mJ/m2 respectively. When calculating the interfacial energy and adhesion work, the surface energy of the E-44 is known to be 41 mJ/m2 [24]. The calculated results obtained from Eqs. (1), (2) and (6) are shown in Tables 2 and 3. The calculation results of surface energy of carbon fibers are shown in Table 2. The surface energy of the original T300 carbon fiber is 46.31 mJ/m2. After deposition of MWCNTs, the surface energy of 0.1 wt% MWCNTs deposited T300 (MWCNTs-T300) is improved obviously. By comparing the surface energy components of T300 before and after MWCNT deposition, the improvement of surface energy of MWCNTs-T300 is mainly due to the increase of the nonpolar component. The nonpolar components of surface energy are caused by the interaction of the electronic dipole and the induced dipole in the adjacent atoms or molecules [25]. Since the surface of the original T300 is smooth (Fig. 1a), the interaction of the dipole is weak, resulting in low nonpolar component of surface energy (31.27 mJ/m2). After the MWCNT deposition, there is a strong p-p stack interaction, which makes the nonpolar component of surface energy increase to 40.71 mJ/m2, and makes MWCNTs-T300 obtain greater surface energy (58.63 mJ/m2). At the same time, the nonpolar component of the surface energy reflects the nonpolar part of the Van der Waals force, which characterizes the morphology of the solid surface. So, MWCNT deposi-
tion increases the surface roughness of carbon fiber, which leads to the improvement of nonpolar component of surface energy. The solid-liquid interfacial energy can reflect the interfacial interaction between the solid and liquid, while the adhesion work is used to measure the energy required for interface separation, which can indicate the bond strength of the solid-liquid interface. Compared with the original T300, the interfacial energy and adhesion work of MWCNTs-T300 and EPs are improved after MWCNT deposition (Table 3). The interfacial energy and adhesion work reached 27.36 mJ/m2 and 72.27 mJ/m2 respectively, which are 49.5% and 4.7% higher than those of the original T300 and EP. This Indicates that the interfacial interaction and interfacial bond strength between MWCNTs-T300 and EP are significantly improved, which will be beneficial to improve the mechanical properties of the carbon fiber/epoxy composite. 3.3. Mechanical properties of carbon fiber/epoxy laminated composites Based on the analysis of the surface morphology, wettability and interfacial properties of MWCNT deposited carbon fiber; the performance of 0.1 wt% MWCNT deposited carbon fiber is the best. Therefore, the MWCNTs-T300/EP laminated composites were prepared by employing 0.1 wt% MWCNT deposited carbon fiber. The mechanical properties of the composites are shown in Table 4. It can be seen that the MWCNTs-T300/EP composites have better comprehensive mechanical properties than the original T300/EP composites.
Table 2 Contact angle of different kinds of droplets on carbon fiber and surface energy of carbon fiber. Surface energy (mJ/m2)
Contact angle (°)
Original T300 0.1 wt% MWCNTs deposited T300
Glycerol
Water
cpS
cdS
cS
45.2 23.8
61.9 51.6
15.04 17.92
31.27 40.71
46.31 58.63
12
C. Xiao et al. / Chemical Physics Letters 703 (2018) 8–16
Table 3 Interfacial energy performance of EP/carbon fiber. 0.1 wt% MWCNTs deposited T300
47.1 ± 0.9 18.30 69.01
40.3 ± 1.7 27.36 72.27
The MWCNTs-T300/EP laminated composites and the original T300/EP laminated composites have the same tensile modulus, which is determined by the tensile modulus of carbon fiber and the fiber content. However, the tensile strength and flexural strength of the 0.1 wt% MWCNTs-T300/EP samples are increased by 17.6% and 15.1% respectively, compared with the original T300/EP samples. Based on the previous analysis of the interfacial properties, the enhancement of interfacial bond strength of MWCNTs-T300 enables the load to transfer from the EP matrix to the carbon fiber more effectively. Therefore, the MWCNTs-T300/ EP samples achieve excellent enhancement effect. Fig. 3 shows the typical tensile stress-strain curves of the original T300/EP laminate composite and the 0.1 wt% MWCNTs-T300/ EP laminate composite. The results show that for the original T300 composites, obvious damage arises under the smaller tensile strain (2.32%), while the tensile stress of the MWCNTs-T300/EP composites decrease significantly under the larger strain (2.68%). This is because: (1) the deposition of MWCNTs improves the interfacial bonding of the MWCNTs-T300/EP composites and is beneficial to the interfacial stress transfer from EP to T300 fiber, thereby enhancing the tensile strength, (2) deposited MWCNTs provide greater interface surface area and interlock at the fibermatrix interface, so the interfacial adhesion is improved, (3) in the process of failure of fiber-matrix interface, MWCNTs on the carbon fiber surface increase the toughness of the EP nearby the interface by pulling out and causing plastic deformation of the EP, which can effectively improve the load transfer efficiency and resistance to strain, so that the shear failure process of interface can dissipate more energy, thus effectively improving the ILSS of the MWCNTs-T300/EP composite. As shown in Table 4, the ILSS of 0.1 wt% MWCNTs-T300/EP laminate is increased by 12.6% compared to the original T300 laminate. In addition, to study the heat resistance of MWCNTs-T300/EP laminated composites, the glass transition temperature (Tg) was measured by DSC. The Tg of the MWCNTs-T300/EP laminated composite is 205 °C, indicating that the composite has good heat resistance. Meanwhile, the Tg of MWCNTs-T300/EP is obviously higher than that of neat EP (181 °C), indicating that the interfacial bonding performance of the MWCNTs-T300/EP composite is good [13,26]. The experimental results show that the properties of MWCNTsT300/EP composites, including the fiber-matrix interface, are greatly improved. In order to study the mechanical properties of MWCNTs-T300/ EP composites under high temperature and humidity conditions, boiling method was used to investigate the temperature and moisture resistance of MWCNTs-T300/EP laminated composite. After boiling, the mechanical properties of 0.1 wt% MWCNTs-T300/EP
700
Tensile Stress / MPa
Contact angle (°) Interfacial energy (mJ/m2) Adhesion work (mJ/m2)
800
Original T300
600 500 400 300 200
Raw CNT-deposited
100 0
Tensile Strain(%) Fig. 3. Tensile stress-strain curves of the original T300 and 0.1 wt% MWCNTs-T300/ EP laminate composites.
laminated composites are shown in Table 5. The mechanical properties of the composites decrease after 24 h of boiling, and the decrease of the flexural strength is larger (retention rate of 84.8%), followed by the ILSS (retention rate of 90.6%), and the flexural modulus decrease little (retention rate >95%), which exhibit a high temperature resistance and excellent moisture resistance. The reason for the slight decrease in ILSS is that during the boiling process, the water gradually penetrates to the interface between the EP and fiber, destroying the force acting on the EP and fiber, and reduces the binding force, therefore the ILSS is slightly reduced. In addition, in the process of water penetration to the interface, some of the water molecules remain in the EP and play a plasticizing effect on the EP matrix, so the flexural strength and flexural modulus of the composite decrease slightly. In general, after 24 h of boiling, it can still maintain high mechanical properties, and the retention rate is around 85% and above, indicating that the MWCNTs-T300/EP composite has good temperature and moisture resistance. 3.4. Fracture morphology analysis of carbon fiber/epoxy laminated composites Fig. 4 shows the fracture morphology of tension face of carbon fiber/epoxy bend specimen. It can be seen that the fracture surface of the MWCNTs-T300/EP composite is smoother than the original T300/EP composite. The original T300/EP composites (Fig. 4a–b) have more holes on the surface (indicated by circles), while MWCNTs-T300/EP composites (Fig. 4c–d) have less, and the surface is smoother. As shown in Fig. 4a–b, many carbon fibers are delaminated from the original T300/EP composite, while the MWCNTs-T300/EP composite has only a few carbon fibers delaminated (Fig. 4c), even without carbon fiber delamination phenomenon (Fig. 4d), which indicates that the fibers of the MWCNTs-T300/EP composite are firmly bonded to the EP resin. Among these fracture surfaces where carbon fibers are
Table 4 Mechanical property of carbon fiber/EP laminated composites. Mechanical property
Flexural strength/MPa
Flexural modulus/GPa
Tensile strength/MPa
Tensile modulus/GPa
ILSS/MPa
Original T300/EP 0.05 wt% MWCNTs-T300/EP 0.1 wt% MWCNTs-T300/EP 0.15 wt% MWCNTs-T300/EP
714 ± 24 742 ± 25 822 ± 27 750 ± 21
41.4 ± 2.1 42.1 ± 2.5 42.7 ± 1.5 42.4 ± 1.8
715 ± 29 743 ± 35 841 ± 34 838 ± 32
93.9 ± 4.8 93.5 ± 4.3 94.0 ± 2.5 92.8 ± 2.7
22.80 ± 1.3 23.69 ± 1.3 25.68 ± 1.5 24.87 ± 1.7
13
C. Xiao et al. / Chemical Physics Letters 703 (2018) 8–16 Table 5 Mechanical properties of MWCNTs-T300/EP laminated composites after 24 h of boiling. Environmental conditions
Flexural strength/MPa
Retention rate/%
Flexural modulus/GPa
Retention rate/%
ILSS/MPa
Retention rate/%
Not boiled Boiled for 24 h
822 ± 27 697 ± 25
84.8
42.7 ± 1.5 40.7 ± 2.2
95.3
25.68 ± 1.5 23.26 ± 1.9
90.6
Fig. 4. Fracture morphology of tension face of carbon fiber/epoxy bend specimen: (a), (b) original T300/EP, (c), (d) MWCNTs-T300/EP.
delaminated, the carbon fiber clusters of the MWCNTs-T300/EP composite (Fig. 4c) are bonded better. The surface area of the compact and non-smooth cross-section after fracture is much larger than that of the flat section, and absorbs more energy during the fracture process, thus improving the strength of the MWCNTsT300/EP composite. In addition, the EP resin is adhered on the surface of the pulled-out carbon fibers in Fig. 4c (indicated by circles), indicating that the interface between the fiber and EP is well bonded. And although there are still a small number of holes on the surface of the specimen, the specimen is not damaged by the holes. The destruction occurs between the fibers and the resins, so the interlaminar shear strength of the MWCNTs-T300/EP composite is higher. Fig. 5 shows the fracture morphology of compression face of carbon fiber/epoxy bend specimen. The fiber fabric under the loading point of bending test are distributed uniformly, the packing density is high, between the fibers is filled with EP, and the EP is not obviously broken. In Fig. 5a–b, the original T300/EP specimen is damaged at the loading point of bending test and results in interlaminar cracks due to the interfacial shear, indicating that the original T300/EP specimen is damaged at the interface between the fiber and EP resin, carbon fiber fabric does not work. For the MWCNTs-T300/EP samples, the crack propagation is perpendicular to the specimen surface (i.e. perpendicular to the carbon fiber fabric), mainly along the fracture of carbon fiber fabric (Fig. 5c–d), rather than the interfacial shear between carbon fiber and EP. This failure mode indicates that MWCNTs enhance the interface by interlocking the fiber-matrix interface [27,28], therefore,
MWCNTs-T300/EP properties.
composite
exhibits
excellent
mechanical
3.5. Interface strengthening mechanism Compared with the original T300 composites, the comprehensive mechanical properties of MWCNTs-T300/EP composites are significantly improved, which is mainly due to the improvement of the interfacial properties between the fiber and the resin. The interface is not only the connecting bridge between the fiber and resin, but also the bond of external load transmits from the matrix to the fiber, the interface structure, properties and its bonding mode will directly affect the destructive behavior of composite. Combined with the above analysis of the interfacial properties, the interface strengthening mechanism of MWCNTs-T300/EP composites can be attributed to the following aspects: (1) local strengthening of the resin matrix near the fiber-matrix interface, (2) Van der Waals force due to the increase of fiber surface area, (3) good surface wettability of EP on MWCNTs deposited carbon fiber, (4) mechanical interlocking of MWCNTs and resin matrix. The interface of the carbon fiber/epoxy composite is not a simple geometric plane, but a three-dimensional interface phase containing the transition region between the fiber and the matrix. According to the surface morphology and roughness of MWCNTs deposited carbon fiber (Fig. 1c), it can be concluded that the Van der Waals force combination and the mechanical interlock increase the interfacial friction of carbon fiber-epoxy matrix and limit the movement of the interface of the composite, which plays an
14
C. Xiao et al. / Chemical Physics Letters 703 (2018) 8–16
Fig. 5. Fracture morphology of compression face of carbon fiber/epoxy bend specimen: (a), (b) original T300/EP, (c), (d) MWCNTs-T300/EP.
important role in the interface enhancement. In addition, surface wettability has a significant influence on the interface bonding of composites. Compared with the original T300 carbon fiber, the contact angle of MWCNTs-T300 carbon fiber and epoxy resin is smaller (40.3°), which indicates that the wettability of MWCNTsT300 is improved effectively, and the interfacial energy and adhesion work of MWCNTs-T300 are improved (27.36 mJ/m2, 72.27 mJ/ m2), so that the bonding performance between MWCNTs-T300 and EP interface is improved, thereby improve the mechanical properties of the MWCNTs-T300/EP composite. The fracture mechanism of carbon fiber/epoxy laminated composites was characterized by SEM. Fig. 6 shows typical SEM images of fracture morphology of carbon fiber/epoxy composites before and after MWCNT deposition. Obviously, no matter whether or not the MWCNTs are deposited, cracks are generated at the interface between the carbon fiber and EP (indicated by the arrows). However, there is a great difference between the fracture surface morphologies of the two composites, especially the failure modes of the interface between carbon fiber and EP on the fracture surface
are obviously different. For the destruction of the original T300 composite (Fig. 6a), the fiber-matrix interface (carbon fiber surface) is substantially clean without adherent of EP (marked by the circle), which indicates the poor interface performance of original T300 and EP result in the easy damage of interface and the low mechanical properties of the original T300/EP composite. Whereas in the fracture surface of MWCNTs-T300/EP composite, the EP resin is obviously retained on the MWCNTs-T300 fibers and is accompanied by the breakage of carbon fibers (as indicated by the circles in Fig. 6b). Therefore, the deposition of MWCNTs on T300 fibers contributes to the mechanical interlocking between the carbon fiber and EP, which improves the interfacial strength and increases the frictional at the interface. As shown in Fig. 7, after the MWCNT deposition, the fracture behavior of the EP near the interface changed significantly. The pull-out of MWCNTs from the fiber surface at the interface (Fig. 7b–c) and the micro plastic deformation of EP near the interface (Fig. 7d) indicates the mechanical interlocking of MWCNTs on the fiber-resin interface and the toughening effect of MWCNTs on
Fig. 6. Fracture morphology of carbon fiber/epoxy composites (a) original T300/EP, (b) MWCNTs-T300/EP.
C. Xiao et al. / Chemical Physics Letters 703 (2018) 8–16
15
Fig. 7. Interface strengthening by MWCNTs, (b), (c) and (d) are partial magnified SEM images of (a).
the EP matrix near the interface. According to the analysis of the interfacial property in the above text, the interfacial bond strength of MWCNTs-T300 and EP matrix is stronger (the adhesion work is 72.27 mJ/m2), and can dissipate more energy in the process of interface failure between MWCNTs-T300 fibers and EP, and thus exhibit excellent toughening effect. The toughening of the EP near the fiber-matrix interface is beneficial to release stress concentration and hinder the formation and propagation of micro-cracks, resulting in interfacial strengthening. Therefore, due to the deposition of MWCNTs, the interface strengthening mechanisms of MWCNTs-T300/EP composites are mainly attributed to interfacial friction and EP matrix toughening. 4. Conclusions (1) The MWCNT deposition was carried out by the aqueous suspension deposition method. The interfacial interaction and interfacial bond strength between MWCNTs-T300 and EP are significantly improved, the contact angle between MWCNTs-T300 and EP was reduced (40.3°), and the interfacial energy and adhesion work are 27.36 mJ/m2 and 72.27 mJ/m2 respectively, which are increased by 49.5% and 4.7% compared with those of the original T300 fiber. (2) After the deposition of MWCNTs, the mechanical properties of MWCNTs-T300/EP laminated composites are effectively improved. The flexural strength is 822 MPa, the tensile strength is 841 MPa and the interlaminar shear strength is 25.68 MPa, which are increased by 15.1%, 17.6% and 12.6% compared with those of original carbon fiber laminated composites respectively. In addition, the MWCNTs-T300/EP composite has good heat resistance and moisture resistance, its Tg is 205 °C. After boiling for 24 h, it can still maintain higher mechanical properties, and the retention rate is around 85% and above.
(3) After the deposition of MWCNTs, the interfacial strength of MWCNTs-T300/EP composites is improved. The interface strengthening mechanisms of MWCNTs-T300/EP laminated composites are the interfacial friction and resin toughening, which is mainly led by the mechanical interlock effect of MWCNTs, the pull-out phenomenon of MWCNTs at the interface and the micro plastic deformation of the EP near the interface. Acknowledgements This work was supported by the National Natural Science Foundation of China, project no. 51708553. It was performed using the equipment at Nanjing University of Aeronautics and Astronautics (NUAA). References [1] S. Chand, J. Mater. Sci. 35 (6) (2000) 1303–1313. [2] H. Qian, A. Bismarck, E.S. Greenhalgh, M.S.P. Shaffer, Compos. Part A-Appl. Sci. 41 (9) (2010) 1107–1114. [3] J. Guo, C. Lu, F. An, S. He, Mater. Lett. 66 (1) (2012) 382–384. [4] T.W. Chou, L. Gao, E.T. Thostenson, Z. Zhang, J.H. Byun, Compos. Sci. Technol. 70 (1) (2010) 1–19. [5] M. Li, Y. Gu, Y. Liu, Y. Li, Z. Zhang, Carbon 52 (2) (2013) 109–121. [6] D. Tasis, K. Papagelis, Carbon nanotube-Polymer Composites, RSC Publishing, 2013. [7] A. Godara, L. Gorbatikh, G. Kalinka, A. Warrier, O. Rochez, L. Mezzo, et al., Compos. Sci. Technol. 70 (9) (2010) 1346–1352. [8] R.J. Sager, P.J. Klein, D.C. Lagoudas, Q. Zhang, J. Liu, L. Dai, et al., Compos. Sci. Technol. 69 (7–8) (2009) 898–904. [9] E.T. Thostenson, W.Z. Li, D.Z. Wang, Z.F. Ren, T.W. Chou, J. Appl. Phys. 91 (9) (2002) 6034–6037. [10] F. An, C. Lu, Y. Li, J. Guo, X. Lu, H. Lu, et al. Mater Design 33(1) (2012) pp. 197– 202. [11] S.D. Faulkner, Y.W. Kwon, S. Bartlett, E.A. Rasmussen, J. Mater. Sci. 44 (11) (2009) 2858–2864. [12] M.A. Bily, Y.W. Kwon, R.D. Pollak, Appl. Compos. Mater. 17 (4) (2010) 347–362.
16
C. Xiao et al. / Chemical Physics Letters 703 (2018) 8–16
[13] A. Warrier, A. Godara, O. Rochez, L. Mezzo, F. Luizi, L. Gorbatikh, et al., Compos. Part A-Appl. Sci. 41 (4) (2010) 532–538. [14] P. Lv, Y.Y. Feng, P. Zhang, H.M. Chen, N. Zhao, W. Feng, Carbon 49 (14) (2011) 4665–4673. [15] K.J. Green, D.R. Dean, U.K. Vaidya, E. Nyairo, Compos. Part A-Appl. Sci. 40 (9) (2009) 1470–1475. [16] Y. Iwahori, S. Ishiwata, T. Sumizawa, T. Ishikawa, Compos. Part A-Appl. Sci. 36 (10) (2005) 1430–1439. [17] X. He, F. Zhang, R. Wang, W. Liu, Carbon 45 (13) (2007) 2559–2563. [18] L. Mei, Y. Li, R. Wang, C. Wang, Q. Peng, X. He, Polym. Polym. Compos. 19 (2) (2011) 107–112. [19] N.A. Siddiqui, M.L. Sham, B.Z. Tang, A. Munir, J.K. Kim, Compos. Part A-Appl. Sci. 40 (10) (2009) 1606–1614.
[20] [21] [22] [23] [24] [25] [26]
S.L. Gao, E. Mäder, R. Plonka, Compos. Sci. Technol. 68 (14) (2008) 2892–2901. S.L. Gao, E. Mäder, R. Plonka, Acta Mater. 55 (3) (2007) 1043–1052. M.E. Schrader, Langmuir 11 (9) (1995) 3585–3589. D.K. Owens, R.C. Wendt, J. Appl. Polym. Sci. 13 (8) (1969) 1741–1747. Weldon, G. Dwight, Pigm Resin Technol 35(2) (2007). F.M. Fowkes, Ind. Eng. Chem. Res. 56 (12) (1964) 40–52. A. Warrier, A. Godara, O. Rochez, et al., Compos. Part A-Appl. Sci. 41 (4) (2010) 532–538. [27] S.S. Wicks, R.G.D. Villoria, B.L. Wardle, Compos. Sci. Technol. 70 (1) (2010) 20– 28. [28] A. Godara, L. Mezzo, F. Luizi, A. Warrier, S.V. Lomov, A.W.V. Vuure, et al., Carbon 47 (12) (2009) 2914–2923.