Science Press
With excellent tensile strength, high stiffness and modulus, low coefficient of thermal expansion, and low density[1-3], the carbon fibres (CFs) have the great potential application to improve the mechanical properties of metal matrix composites as the reinforcement[4]. CFs reinforced copper matrix (CF/Cu) composites have the advantages of good electrical and thermal conductivity (due to Cu) and excellent mechanical properties (due to CFs), and display superior mechanical and physical properties, including the low coefficient of thermal expansion, high specific strength and modulus, and good wear resistance, therefore attracting widespread attention in the manufacture of integrated structure/function materials, such as electrical brushes and pantograph slides[5,6]. However, the mechanical properties of CF/Cu composites are worse than their theoretical values because of the imperfect load transfer effect, which is caused by the poor wettability and the absence of interfacial reaction between CFs and Cu matrix[7,8]. Thus, improving the interfacial wettability and interfacial bonding strength between CFs reinforcement and Cu matrix is important for the property enhancement of the CF/Cu composites[5,7,9-12].
The CF/Cu composites prepared by sintering and electroplating methods exhibit the better defect-free interface, compared with the ones prepared by the mixing and sintering methods. However, the weak mechanical interface bonding is barely ameliorated with the preparation method change[8]. In order to enhance the interfacial bonding strength, Veillere et al[13] studied the effects of matrix alloying with chromium on the interface type and bonding nature, and found that the bonding strength is increased and the interface bonding type changes from mechanical bonding to metallurgical bonding due to the formation of chromium carbide (Cr3C2) interphase at the interface. Whereas, the reaction between CFs and chromium results in the graphitization and the strength reduction of CFs, thereby leading to the ripping shape of the fracture of CFs. Yang et al[6] found that the titanium and nickel coatings deposited on CFs surface can effectively improve the interface bonding strength between CFs and Cu matrix and avoid the pore defects. It is worth noting that Ni is the optimal candidate for interface modification of C/Cu composites, because it can not only dissolve into Cu infinitely[14], but also has the fine wettability with carbon. Compared with other metal elements, Ni has obviously better wettability, owing to the Ni-C-O bonds formed between CFs and Ni[15] through the chemical solution deposition method. Ni alloy coatings not only increase the density of C/Cu composites, but also change the crack propagation mechanism[16,17], which indirectly enhances the load-bearing capacity of composites. Unfortunately, the graphitization of CFs caused by the diffusion-precipitation between Ni and C during sintering process of composites may decrease the tensile strength of CFs. Dong et al[18] studied the effect of Ni-Cu duplex alloy coatings on the friction and wear resistance of CF/Al composites, and found that CF graphitization induced by Ni can be restrained due to the presence of NiAl compound, which hinders the formation of Cu-Ni solid solution, crack initiation, and crack propagation at the interface. Currently, the effects of Al-M alloy coatings on interface microstructure of CF/Cu composites are rarely reported.
Co has excellent wettability and compatibility with CFs[19] and may restrain the CF graphitization through the alloying on nickel coatings. In this research, the nickel-cobalt-phosphorus (Ni-Co-P) coated CFs were prepared by palladium-free electroless plating, and then the Cu matrix composites with different contents of Ni-Co-P coated CFs and uncoated CFs were synthesized by vacuum hot pressing sintering. By the comparison of stress-strain curves of the copper matrix composites reinforced by Ni-Co-P coated CFs, namely CF/Cu(Ni-Co-P) composites, the copper matrix composites reinforced by uncoated CFs, namely CF/Cu composites, and the copper matrix composites reinforced by Ni coated CFs, namely CF/Cu(Ni) composites, the influence of Ni-Co-P coatings and CFs on tensile properties of Cu matrix composites was analyzed. In addition, the strengthening effect of reinforcement and fracture mechanism of the composites were studied.
The essential parameters of original CFs (Gansu Haoshi Carbon Fibre Co., Ltd, Gansu, China) are shown in Table 1. The composition of the electroless plating solution for Ni-Co-P coating is listed in Table 2, and all chemical reagents used in the electroless plating process are analytically pure.
Table 1
Essential parameters of CFs
| Specification | Diameter, D/μm | Elastic modulus, E/GPa | Elongation, δ/% | Density, ρ/g‧cm-3 | Carbon content/wt% |
|---|
|
PAN-T300 |
~7 |
206 |
1.5 |
1.76 |
98.5 |
Table 2
Composition and functions of electroless plating solution for Ni-Co-P alloy coating
| Component | Content/g‧L-1 | Function |
|---|
|
NiSO4·6H2O |
13.15 |
Main salt |
|
CoSO4·7H2O |
14.05 |
Main salt |
|
C6H5Na3O7 |
40.00 |
Complexing agent |
|
(NH4)2SO4 |
40.00 |
Buffer |
|
NaH2PO2 |
20.00 |
Reducing agent |
|
NaOH |
- |
pH adjustment |
Fig.1 shows the preparation process of Ni-Co-P coating on CF surface. Firstly, the initial CFs were immersed in 68wt% HNO3 at 90 °C for 30 min, subsequently washed several times by the distilled water, and then dried in thermostatic drying chamber at 60 °C for 6 h. Secondly, the CFs fixed on the square framework were immersed in 2 mol/L NiSO4 solution for 10 min, and then dried at 100 °C for 4 h. Thirdly, the activated CFs washed by distilled water were reduced in the reduced aqueous solution containing 20 g/L KBH4 and 50 g/L NaOH at ambient temperature for 10 min, and were subsequently rinsed by the distilled water to clean the residual reductant. Finally, the CFs after activation-reduction were immersed in Ni-Co-P electroless plating solution at 70 °C for 10 min, and then rinsed and dried at 40 °C for 8 h using the vacuum drying oven.
Fig.1 Preparation process of Ni-Co-P coating on CF surface
As shown in Fig.2, the CF/Cu(Ni-Co-P) composites were firstly fabricated through the repetitive overlapping of copper foil with the thickness of 0.01 mm and Ni-Co-P coated CFs. The refined copper foil was cut into the foils of 12 mm×50 mm, and the Ni-Co-P coated CFs were uniformly dispersed into the lamellar structure. Subsequently, the lamellar fibres and Cu foil were placed layer by layer in the graphite mold. The CFs content was controlled as 10vol%, 20vol%, 30vol%, and 40vol% in the composites. Ultimately, the CF/Cu composites were sintered by vacuum hot pressing at 850 °C and 30 MPa for 40 min. The same process was used to prepare the CF/Cu composites with 30vol% CFs and the CF/Cu(Ni) composites with 30vol% CFs.
Fig.2 Preparation of CF/Cu composites
X-ray diffraction (XRD) analysis of phase composition of Ni-Co-P coatings was conducted by the D/max 2400 XRD with the monochromatized Cu Kα radiation (λ=0.154 178 nm) at a scanning speed of 5°/min. The microstructure and element content of Ni-Co-P coated CFs and the distribution of fibres and interfacial elements of Cu matrix composites were investigated by the transmission electron microscope (TEM) and the scanning electron microscope (SEM, Quanta FEG-450 SEM) coupled with the energy dispersive spectrometer (EDS). The tensile tests for composites were conducted by the electronic universal testing machine (MTS-E44.304, USA) with a constant tensile rate of 0.2 mm/min at ambient temperature. Before the tensile test, the Cu matrix composites were firstly processed into the tensile specimens with a gauge distance of 15 mm by the electrical discharge machining, and were subsequently polished with 2000# waterproof abrasive paper. The fracture morphologies of the composites were observed by SEM.
2.1 XRD analysis of CFs before and after modification
XRD patterns of CFs with and without Ni-Co-P coatings are shown in Fig.3. It can be seen that the uncoated CFs have two diffraction peaks at 2θ=25.8° and 42.6°, corresponding to the (002) and (100) crystal planes of amorphous carbon, respectively. Moreover, some small diffraction peaks appear, corresponding to the graphitization crystallization of the trace amount of carbon. Compared with the uncoated CFs, the Ni-Co-P coated CFs exhibit the crystallization peaks of Ni and Co with the face-centered cubic structure, and no diffraction peak of carbon exists[20]. Some diffraction peaks suggest that the P element exists in the form of nickel phosphorus compounds, such as Ni7P3 and Ni2P phases, which are formed by the Ni-P codeposition. Besides, it is also confirmed that the coatings are fairly pure and have high crystallinity.
2.2 Morphology and deposition of Ni-Co-P coatings
SEM morphologies of CFs with and without Ni-Co-P coatings are shown in Fig.4. According to Fig.4a, a number of axial ditches appear on the surface of uncoated CFs after acid treatment, which is conducive to the adsorption of activated Ni2+ ions and can improve the mechanical bonding strength between the coatings and CFs. After deposition at 70 °C for 10 min, the smooth and consecutive Ni-Co-P coatings with a homogeneous thickness of ~0.7 μm are obtained on CFs surface (Fig.4c). However, there are some cellular Ni-Co-P particles distributed on the coating surface. According to the EDS spectrum in Fig.4c, the Ni content of the coating is 61.4wt%, which is significantly higher than the Co content (~28.9wt%) and P content (~9.7wt%). The microstructure and composition of Ni-Co-P coatings are determined by the differences between the Ni and Co deposition mechanisms. After the reduction by the KBH4+NaOH mixed solution, the Ni2+ ions adsorbing on CFs surface are reduced to active Ni atoms (Ni0). According to the atomic hydrogen theory[21], the active hydrogen Had derives from the reaction of sodium hypophosphite with the catalysis of Ni0, as follows:
Fig.4 SEM morphologies of CFs without (a) and with Ni-Co-P coatings after deposition for 3 min (b) and 10 min (c)
Ni2+ and Co2+ react with Had to form Ni and Co, as shown in Eq.(2) and Eq.(3), respectively:
The reaction rate of Eq.(2) is faster than that of Eq.(3) owing to the relatively high standard electrode potential of Ni (EΘ=-0.250 V), compared with that of Co (EΘ=-0.277 V) [22]. Thus, the deposition amount of Ni atoms is greater than that of Co atoms, i.e., the Ni-Co-P coatings are mainly composed of Ni and a small amount of Co and P[23]. Besides, the reduced Ni and Co atoms are not deposited on CFs surface. They gather together to form the atom clusters of different sizes at the solid-liquid interface[24,25] and subsequently to form the Ni-Co-P particles with chain shape, as shown in Fig.4b. The Ni-Co-P particles with different sizes are formed in the initial stage of deposition (3 min). After deposition for 10 min, the continuous Ni-Co-P coatings with trace cellular particles appear and cover the CFs surface.
2.3 Microstructure of CF/Cu(Ni-Co-P) composites
The microstructures of CF/Cu(Ni-Co-P) composites with 10vol%, 20vol%, 30vol%, and 40vol% CFs are shown in Fig.5. The CFs are uniformly distributed in the Cu matrix and show tight bonding with the Cu matrix. No agglomeration occurs when CFs content is below 30vol%. With increasing the CFs content to 40vol%, the homogenous distribution of CFs becomes worse and a few cavity defects occur at the interface caused by the excess addition of fibres, as shown in Fig.5d. The load transfer effect and the forming ability of Cu matrix are thereby worsened. In addition, Ni, Co, and P elements are accumulated on the interface between the fibres and Cu matrix owing to the non-diffusion effect of the Ni-Co-P coatings. This result indirectly indicates that the composite strength is unrelated to the alloying strengthening effects of the alloy coatings.
Fig.5 Microstructures of CF/Cu(Ni-Co-P) composites with 10vol% (a), 20vol% (b), 30vol% (c), and 40vol% (d) CFs
TEM microstructures of the interfaces in CF/Cu(Ni-Co-P) composites are shown in Fig.6. Fig.6b shows that there is an obvious boundary between CFs and coatings and no cavities occur, indicating the good interface bonding state. Besides, the microstructures of CFs near the interface show no significant alteration, suggesting that the Ni-Co-P coatings do not lead to the graphitization of fibres during the high temperature sintering, i.e., the tensile properties of CFs are not influenced by the Ni-Co-P coatings. However, the interface boundary between coatings and Cu matrix (Fig.6c) is not obvious because of the interface reaction. The metallurgical bonding interface can ensure the effective transfer of load from Cu matrix to fibres.
Fig.6 TEM microstructure of interfaces in CF/Cu(Ni-Co-P) composites (a); magnified TEM microstructures of area A (b) and area B (c) in Fig.6a
2.4 Effects of CFs content and different coatings on tensile properties of composites
The ultimate tensile strength (UTS), yield strength (YS), and elongation (EL) of CF/Cu(Ni-Co-P) composites with 10vol%, 20vol%, 30vol%, and 40vol% CFs fabricated by the vacuum hot pressing at 850 °C and 30 MPa for 40 min are shown in Fig.7. With increasing the CFs content, UTS and YS are firstly increased and then decreased. When CFs content is 30vol%, the CF/Cu(Ni-Co-P) composites have the highest UTS of ~415 MPa and YS of ~275 MPa, which are increased by 157.8% and 212.5%, compared with those of the copper matrix (UTS=161 MPa, YS=88 MPa), respectively. In addi-tion, UTS and YS of CF/Cu(Ni-Co-P) composites with 40vol% CFs are 360 and 230 MPa, respectively, which are lower than those of the CF/Cu(Ni-Co-P) composites with 30vol% CFs.
Fig.7 Tensile properties of CF/Cu(Ni-Co-P) composites with different CFs contents
For the composites reinforced by continuous fibres, the tensile properties are determined by the fibre strength and the fibre content, which can be predicted based on the law of mixtures, as expressed by Eq.(4) [26], as follows:
where σc and σf are UTS of the composites and fibres, respectively; Vf and Vm refer to the volume fraction of fibres and matrix, respectively; Em and εf represent the elastic modulus of matrix and the fracture strain of fibres, respec-tively. Therefore, the tensile properties of the composites are increased with increasing the CFs content. However, when CFs content exceeds 30vol%, Eq.(4) is no longer suitable for the prediction of tensile properties of CF/Cu(Ni-Co-P) composites. The cavity defects in CF/Cu(Ni-Co-P) composites with 40vol% CFs may act as the starting points for the crack initiation and propagation, thereby leading to the fibre debonding and pullout, which reduces the load transfer efficiency of Cu matrix. Additionally, the measured tensile strength of the composite is far lower than the theoretical strength. There are two main reasons for this result. Firstly, the residual strength of CFs is decreased because of the partial diffusion between Ni and fibres during the hot pressing sintering[9], so it is less than their room-temperature strength. Secondly, the inhomogeneous loading capacity caused by the thermal expansion mismatch between fibres and matrix results in the failure of fibres, i.e., CFs cannot bear the load at the same time, therefore reducing the material strength.
Fig.8 shows the tensile properties of Cu, CF/Cu, CF/Cu(Ni), and CF/Cu(Ni-Co-P) composites to investigate the effect of Ni-Co-P coatings on composites. It can be seen that YS and UTS of pure Cu are 88 and 160 MPa, respectively, which are much lower than those of other composites, although its EL is the highest among these composites. After the reinforcement with CFs, the tensile properties of CF/Cu(Ni), and CF/Cu(Ni-Co-P) composites are higher than those of the CF/Cu composite. Compared with those of CF/Cu composite, UTS (415 MPa) and YS (275 MPa) of CF/Cu(Ni-Co-P) composite are increased by 53.7% and 50.5%, respectively; while UTS (303 MPa) and YS (240 MPa) of CF/Cu(Ni) composite are increased by 12.2% and 30.4%, respectively. This result demonstrates that the alloy coatings have a better interface modification effect on the composite properties than the single-element metallic coatings do.
Fig.8 Tensile properties of Cu, CF/Cu, CF/Cu(Ni), and CF/Cu(Ni-Co-P) composites
2.5 Fracture mechanism of Cu matrix composites
The stress-strain curves of Cu, CF/Cu(Ni-Co-P), CF/Cu(Ni), and CF/Cu composites are shown in Fig.9a. According to the stress-strain curve of pure Cu, there is no obvious yielding point, and the wide plastic deformation zone indicates that the pure Cu possesses excellent plasticity and toughness. After the reinforcement with different CFs, the plastic deformation zones of composites have different degrees of shortening, which suggests that the failure modes of Cu matrix composites change and are determined by the interface bonding states based on the coating status. From the stress-strain curve of CF/Cu composite, the composites show linear elastic deformation when the stress is below 40 MPa. When
Fig.9 Stress-strain curves of pure Cu and Cu matrix composites (a); fracture morphologies of CF/Cu (b), CF/Cu(Ni) (c), and CF/Cu(Ni-Co-P) (d) composites
the stress increases to ~40 MPa, the microcracks begin to initiate at the interface and are propagated to the Cu matrix, leading to the variation of deformation mode from linear elastic deformation to nonlinear plastic deformation. According to the fracture morphology of CF/Cu composite (Fig.9b), the composite damage mainly includes three types:interface debonding, fibres fracture, and sliding. Thus, the failure mode is an accumulative fracture. For the CF/Cu(Ni) composite, the stress-strain curve has a similar tendency with that of CF/Cu composite, suggesting that the essential failure mode of composites does not change. According to the fracture morphology of CF/Cu(Ni) composite (Fig.9c), the fibre pull-out length is shorter than that in the CF/Cu composite, which indirectly proves that the Ni coating can enhance the interface bonding strength and restrain the microcrack initiation and propagation, therefore significantly improving the tensile strength of composites and the critical stress of deformation. However, there are obvious interfacial debonding between the fibres and matrix, inferring that the failure mode of CF/Cu(Ni) composite is still the accumulative fracture. The elastic modulus of CF/Cu(Ni-Co-P) composite is similar to that of CF/Cu and CF/Cu(Ni) composites in the initial loading stage. With increasing the stress to ~275 MPa, the loading exceeds the UTS of partial fibres, so the defor-mation of composites changes from linearity to nonlinearity caused by the initiation and propagation of microcracks on fibres. The fibre debonding is restrained by the strong interface bonding strength between CFs and Cu matrix. With further increasing the loading, the fracture of partial fibres decreases not only the strength of fibres, but also the strength of matrix due to the crack propagation. These phenomena can decrease the elastic-plastic modulus with increasing the strain. As shown in Fig.9d, the majority of the fibres and matrix break on the same plane at the same time, and only a few fibres are slightly pulled out. Thus, the failure mode of CF/Cu(Ni-Co-P) composite is the non-accumulation fracture.
1) Smooth and continuous Ni-Co-P coatings containing 61.4wt% Ni, 9.7wt% P, and 28.9wt% Co with the thickness of ~0.7 μm can be obtained on carbon fibres (CFs) by palladium-free electroless plating at 70 °C for 10 min.
2) The copper matrix composites reinforced by Ni-Co-P coated CFs, namely CF/Cu(Ni-Co-P) composites, with CFs re-inforcement content of 10vol%, 20vol%, 30vol%, and 40vol% can be fabricated by the vacuum hot pressing sintering at 850 °C for 40 min. The tensile properties of CF/Cu(Ni-Co-P) compos-ites are increased firstly and then decreased with increasing the CFs content. The CF/Cu(Ni-Co-P) composites with 30vol% CFs have the highest ultimate tensile strength of ~415 MPa and yield strength of ~275 MPa, while their elongation is inferior.
3) The failure mode of CF/Cu(Ni-Co-P) composites is the non-accumulation fracture, whereas that of other Cu matrix composites reinforced by CFs is the accumulation fracture.
References
1 Yao Jinjin, Chu Di, Han Yanqiu et al. Advanced Materials Research[J], 2013, 850-851: 173 [Baidu Scholar]
2 Shirvanimoghaddam K, Hamim S U, Akbari M K et al. Composites Part A: Applied Science and Manufacturing[J], 2017, 92: 70 [Baidu Scholar]
3 Newcomb B A. Composites Part A: Applied Science and Manufacturing [J], 2016, 91(1): 262 [Baidu Scholar]
5 Yang Ping, You Xin, Yi Jianhong et al. Journal of Alloys and Compounds[J], 2018, 752: 431 [Baidu Scholar]
6 Yang Yana, Chen Wenge, Xue Yuanlin. Chinese Journal of Materials Research[J], 2021, 35(6): 467 (in Chinese) [Baidu Scholar]
7 Liu Lei, Tang Yiping, Zhao Haijun et al. Journal of Materials Science[J], 2008, 43(3): 974 [Baidu Scholar]
8 Ouyang Wenjing, Jia Jiangang, Ma Qin et al. Acta Materiae Compositae Sinica[J], 2016, 33(12): 2824 (in Chinese) [Baidu Scholar]
9 Jia Jiangang, Gao Changqi, Liu Diqiang et al. Materials Reports[J], 2018, 32(14): 2462 (in Chinese) [Baidu Scholar]
10 Wan Y Z, Wang Y L, Luo H L et al. Materials Science and Engineering A[J], 2000, 288(1): 26 [Baidu Scholar]
11 Xiao Peng, Jiang Xu, Zhu Jiamin et al. Rare Metal Materials and Engineering [J], 2019, 48(10): 3265 (in Chinese) [Baidu Scholar]
12 Liu Xueran, Liu Yongbing, Sun Yu et al. Rare Metal Materials and Engineering[J], 2012, 41(S2): 769 (in Chinese) [Baidu Scholar]
13 Veillere A, Sundaramurthy A, Heintz J M et al. Acta Materialia[J], 2011, 59(4): 1445 [Baidu Scholar]
14 Zhou Qing, Huang Ping, Liu Mabao et al. Journal of Alloys and Compounds[J], 2017, 698: 906 [Baidu Scholar]
15 Hua Zhongsheng, Liu Yihan, Yao Guangchun et al. Journal of Materials Engineering and Performance[J], 2012, 21(3): 324 [Baidu Scholar]
16 Dong Ruifeng, Cui Zhenduo, Zhu Shengli et al. Metallurgical and Materials Transactions A[J], 2014, 45(11): 5194 [Baidu Scholar]
17 Zhu Z, Kuang X, Carotenuto G et al. Journal of Materials Science[J], 1997, 32(4): 1061 [Baidu Scholar]
18 Dong Shuhan, Yuan Huiyong, Cheng Xiaochao et al. Metals[J], 2020, 10(11): 1542 [Baidu Scholar]
19 Yi Pei, Chen Biankun, Wu Yunshu. Acta Materiae Compositae Sinica[J], 1987, 4(3): 57 (in Chinese) [Baidu Scholar]
20 Chow S L, Hedgecock N E, Schlesinger M et al. Journal of the Electrochemical Society[J], 1972, 119(12): 1614 [Baidu Scholar]
21 Jiang Xiaoxia, Shen Wei. The Fundamentals and Practice of Electroless Plating[M]. Beijing: National Defense Industry Press, 2000 (in Chinese) [Baidu Scholar]
22 Xuan Tianpeng, Zheng Xiaohua, Deng Zonggang. Materials Protection[J], 1997, 30(1): 20 (in Chinese) [Baidu Scholar]
23 Narayanan T S N S, Selvakumar S, Stephen A. Surface and Coatings Technology[J], 2003, 172(2-3): 298 [Baidu Scholar]
24 Xu Chaoxing, Yang Wenbin, Xie Changqiong et al. Journal of Southwest University of Science and Technology[J], 2013, 28(1): 13 (in Chinese) [Baidu Scholar]
25 Sun Hongfei, Wu Yiyong, Shao Qian et al. The Chinese Journal of Nonferrous Metals[J], 2000, 10(1): 81 (in Chinese) [Baidu Scholar]
26 Cheng Laifei, Mei Hui, Liu Yongsheng et al. Principle and Technology of Composite Materials[M]. Xi'an: Northwest Poly-technical University Press, 2016 (in Chinese) [Baidu Scholar]
