- Email: [email protected]
carbon composite by controlling carbon nanotube growth position in carbon felt
Materials Science & Engineering A 564 (2013) 71–75
Contents lists available at SciVerse ScienceDirect
Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea
Rapid communication
The reinforcement and toughening of pyrocarbon-based carbon/carbon composite by controlling carbon nanotube growth position in carbon felt Qiang Song a, Ke-zhi Li a,n, Le-hua Qi b, He-jun Li a, Jin-hua Lu a, Lei-lei Zhang a, Qian-gang Fu a a b
State Key Laboratory of Solidification Processing, Northwestern Polytechnical University, Xi’an 710072, China School of Mechanical Engineering, Northwestern Polytechnical University, Xi’an 710072, China
a r t i c l e i n f o
abstract
Article history: Received 18 July 2012 Received in revised form 19 November 2012 Accepted 21 November 2012 Available online 29 November 2012
The flexural properties of pyrocarbon-based carbon/carbon (C/C) composites reinforced by in situ grown carbon nanotubes are reported. The effect of nanotube growth position (on carbon fibers or in the spaces between them) was investigated. A method of reinforcing and toughening C/C composites is proposed and the corresponding mechanisms are discussed. Crown Copyright & 2012 Published by Elsevier B.V. All rights reserved.
Keywords: Mechanical characterization Composites Fracture
Fig. 1. SEM images of CNTs with different growth positions: (a) CNTs grown on carbon fibers and (b) CNTs grown in the spaces between fibers.
1. Introduction Over the past few decades, much different effort has been devoted to improve the toughness of carbon/carbon (C/C) composites [1–3]. An efficient work is adding nano-filler, such as carbon nanofiber (CNF), into the composites [2,3]. In the cases, resin or pitch carbon was used as the matrix and the toughening mechanisms have been found to be ‘‘CNF bridging’’ and ‘‘CNF pull-out’’. However, for C/C
n
Corresponding author. Fax: þ86 29 8849 5764. E-mail address: [email protected] (K.-z. Li).
composites with pyrocarbon (PyC) matrix, the existing reports show that adding nano-fillers such as carbon nanotube (CNT) always leads to the brittle fracture of the composites despite a largely increased mechanical strength [4–6]. The reasons for the brittle fracture are various.One of them is the failure of the toughening mechanisms mentioned above because the bonding between CNT and PyC is very strong [7] that inhibits the long-distance pull-out of CNT from carbon matrix. Another important one is the overly-enhanced F/M interface bonding because of the introduction of CNTs onto carbon fibers which blocks the fiber pull-out and decreases the fracture toughness [5,6], even the CNT content is very low [5].
0921-5093/$ - see front matter Crown Copyright & 2012 Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.msea.2012.11.074
72
Q. Song et al. / Materials Science & Engineering A 564 (2013) 71–75
Recently, it is found that carbon matrix with multilayer microstructures is beneficial to the mechanical properties of C/C composites, especially the toughness [8,9]. On the other hand, introducing
CNTs into the matrix area of C/C composites can induce the formation of PyC matrix with multilayer microstructures during chemical vapor deposition (CVD) [9,10], under the precondition of not increasing the F/M interface bonding. Thus, it is very possible to develop a novel method of reinforcing and meanwhile toughening PyC-based C/C composites by adding CNTs into the space between carbon fibers rather than onto fibers. In the present work, this issue was demonstrated based on the investigation into the effects of CNT growth position (on the fibers or in the spaces between them) on the flexural properties of C/C composites.
2. Experimental
Fig. 2. Representative flexural stress–strain curves of the three composites.
Table 1 Mechanical properties of the three kinds of C/C composites. Composites CNT content (wt.%)
CNT Bulk length(mm) density (g m-3)
Flexure strength (MPa)
FD
C/C CNTP1-C/C
0 2–4 4–20 8–20 5–18 5–20
143 75 182 78 251 77 179 710 196 712 212 78
0.27 7 0.07 0.037 0.01 o0.01 0.56 7 0.10 0.58 7 0.08 0.49 7 0.10
CNTP2-C/C
0 2.0 9.5 1.9 6.2 10.3
1.55 1.57 1.56 1.57 1.57 1.56
Carbon felts (bulk density: 0.43 g/cm3; fibers diameter: 79 mm) were used as preform materials. The growth of CNTs in carbon felts was accomplished by catalytic CVD using Ni(NO3)2 6H2O as catalyst precursor. Incipient wetness technique was adopted to introduce catalyst into felts using acetone as solvent. The gas system for CNT growth was CH4/N2 (flow ratio: 1/10). The growth temperature was 1020 1C. The growth time was 1 h. To obtain CNTs grown on carbon fibers (named CNTs-P1), before dipped in catalyst solution, carbon felts were firstly coated with a layer of PyC (several dozen nanometers) to prevent the dissolution of metal catalyst into carbon fibers and then were treated in 65 vol.% HNO3 at 60 1C for 10 h. To obtain CNTs grown in the spaces between fibers (named CNTs-P2), carbon felts were directly dipped in catalyst solution. After CNT growth, all the felts were densified in a CVD reactor under rough laminar (RL) PyC deposition conditions [11]. For comparison, C/C composites with the same felts were prepared under the same CVD conditions. The flexural properties of the three composites were tested by three-point-bending test at room temperature, conducted at loading speed of 0.5 mm/min and support span of 20 mm length.
Texture of RL PyC
Fig. 3. SEM microstructure and its model of the three composites: (a) C/C composites; (b) CNTP1-C/C composites; (c) CNTP2-C/C composites (the inset shows that RL PyC is deposited nearly around a single CNT).
Q. Song et al. / Materials Science & Engineering A 564 (2013) 71–75
The size of bending specimens was 35 mm 5 mm 2 mm. The number of specimens used in the flexural test was no less than five for every test.
3. Results and discussion Scanning electron microscopy (SEM) image (Fig. 1a) shows CNTs-P1, with straight body and length of 2 4 mm, have
73
approximately radial growth orientation (relative to carbon fibers), forming a dense three dimensional (3D) network structure (Fig. 1a inset). CNTs-P2 (Fig. 1b) have long length (5 18 mm) and a scattered distribution in the spaces between fibers. CNTs-P1 and CNTs-P2 account for 2.0 wt.% and 1.9 wt.% of the hybrid felt, respectively. Fig. 2 shows the representative flexural stress–strain curves of the three composites. After adding CNTs-P1 and CNTs-P2, similar mechanical strength improvements are obtained and the flexural
Fig. 4. SEM images of fracture surfaces for C/C composites (a, d), CNTP1-C/C composites (b, e), CNTP2-C/C composites (c, f).
74
Q. Song et al. / Materials Science & Engineering A 564 (2013) 71–75
Fig. 5. Enlarged SEM images of the fractured CNTP2-C/C composites: (a) shows the rough surface of fractured PyC matrix, in which interfaces between CNT/PyC units and fractured CNT/PyC structures can be seen; (b) shows a fractured CNT/PyC structure with step-like shape and some delaminated RL PyC sheets around it.
strength increases by 27% and 25%, respectively (Table 1). However, the fracture behaviors of the two composites are greatly different. The composites reinforced by CNTs-P2 (named CNTP2C/C composites) and C/C composites exhibit pseudo-plastic fracture behaviors [11]; while, C/C composites reinforced by CNTs-P1 (CNTP1-C/C composite) shows a typically brittle fracture. Quantitative comparison is done based on the calculation of ductility factor FD (which can estimate the quasi-ductile fracture behavior of C/C composites [11]). As shown in Table 1, CNTP2-C/C composites have over 1700% and 107% FD improvements, respectively compared with CNTP1-C/C and C/C composites. The great differences between the fracture behaviors of the two CNT-hybridized composites can be explained by their greatly different microstructures. Without CNTs, RL PyC is deposited layer by layer around carbon fiber (Fig. 3a). After adding CNTsP1, PyC with very random orientation (i.e. isotropic (ISO) PyC) is deposited nearly around fibers (Fig. 3b) though the deposition conditions should lead to RL PyC. Only in the area away from CNTs, layered PyC is formed. The formation of ISO PyC is related to the very disorderly internal structure of the 3D CNT network [5]. ISO structures, which are seriously intertwisted, can immensely inhibit the deflection of matrix microcracks in F/M interface area and then lead to the brittle fracture of CNTP1-C/C composites [12] though can increase the mechanical strength obviously [9]. More importantly, the strongly enhanced F/M interface bonding due to the radially grown CNTs (Fig. 4e inset, CNT pull-out can be seen), characterized by the absent fiber pullout (Fig. 4b and e), further leads to the high strength and the poor ductility of CNTP1-C/C composites. After adding CNTs-P2, RL PyC is deposited around CNT (Fig. 3c inset) rather than fiber in matrix area and CNT/RL unit is formed; while, in F/M interface area, RL PyC is still deposited around carbon fibers (labeled by dotted arrows in Fig. 3c). Due to the similar microstructure of F/M interface, the fiber pull-out situation of fractured CNTP2-C/C composites is similar to that of fractured C/C composites (Fig. 4c, f and 4a, d). Compared with C/C composites, the obviously increased mechanical strength and largely improved toughness of CNTP2-C/C composites should be attributed to multilayer carbon matrix. Concretely, CNT/RL units and the interfaces between them constitute multilayer structures, which lead destructive cracks to spread along multiple paths [8] (Fig. 4f shows a rough fracture surface of PyC matrix of CNTP2-C/C composites; while, the fracture surface of C/C composites shows very flat, Fig. 4d inset). More energies are dissipated during this course, which can increase mechanical strength and also make flexural stress release gently [13], leading to sliding regions occurred in stressstrain curve which corresponds to an increased FD. The multiple
paths mentioned above mainly include the interfaces between CNT/RL units (Fig. 5a) and CNT/RL unit itself (Fig. 5b shows the step-like fracture of a CNT/PyC unit and the delaminated RL PyC sheets around it). In addition, CNT pull-out has also been found (Fig. 5b); however, the quantity of pulled-out CNTs is very small and the length is also very short (100–500 nm). In contrast, for CNT-reinforced pitch-based C/C composites, the quantity of pulled-out CNTs is considerable and the pull-out length ranges from 3 mm to over 5 mm [3]. Thus, it can be said that the contribution of CNT pull-out to the high strength and the high ductility must be very small. In our work, it is also found that increasing the mass content of CNTs-P2 can further reinforce and toughen C/C composites, as shown in Table 1. When this content is over 10 wt.%, the ductility of the composites will decrease, however, the FD value is still much larger than that of CNTP1-C/C and C/C composites. On the other hand, increasing the mass content of CNTs-P1 can further lower the ductility, though which can also endow a very high flexural strength to C/C composites. Anyway, the significance of these results consists in the fact that adding CNTs into matrix is indeed an effective method of reinforcing and also toughening C/C composites.
4. Conclusions In this work, the effects of CNT growth position on the flexural properties of PyC-based C/C composites were investigated. Results show that adding CNTs in the spaces between carbon fibers can not only enhance the mechanical strength but also drastically improve the fracture toughness of C/C composites, compared with adding CNTs on fibers. Different from the previous work [2,3], the excellent mechanical properties benefit from the multilayer structured PyC matrix formed under the influence of CNTs dispersed in the spaces between fibers rather than the roles of CNT bridging and CNT pull-out during the fracture of C/C composites.
Acknowledgments This work has been supported by the program of introducing talents of discipline to universities (Grant no. B08040) and the Ph.D. thesis innovation fund of NWPU (cx201214), key grant project of Chinese Ministry of Education (313047) and National Natural Science Foundations of China (51275417, 50832004 and 51202194).
Q. Song et al. / Materials Science & Engineering A 564 (2013) 71–75
References [1] G. Chollon, O. Siron, J. Takahashi, H. Yamauchi, K. Maeda, K. Kosaka, Carbon 39 (2001) 2065–2075. [2] H.L. Liu, Z.H. Jin, Z.B. Hao, X.M. Zeng, Mater. Sci. Eng. A 483–484 (2008) 316–318. [3] H.L. Li, H.J. Li, J.H. Lu, K.Z. Li, C. Sun, D.S. Zhang, Mater. Sci. Eng. A 547 (2012) 138–141. [4] P. Xiao, X.F. Lu, Y.Q. Liu, L.L. He, Mater. Sci. Eng. A 528 (2011) 319–323. [5] K.Z. Li, Q. Song, L.Y. Wu, H.J. Li, Q.J. Gong, J. Inorg. Chem. 27 (2011) 1001–1007. [6] Q. Song, K.Z. Li, H.L. Li, H.J. Li, C. Ren, Carbon 50 (2012) 3949–3952. [7] H. Allouche, M. Monthioux, Carbon 43 (2005) 1265–1278.
75
[8] H.J. Li, H.L. Li, K.Z. Li, Y.J. Wang, D.S. Zhang, J.H. Lu, J. Mater. Sci. 46 (2011) 4667–4674. [9] K.Z. Li, H.L. Deng, H.J. Li, X. Li, X.N. Zhao, H.J. Luo, Mater. Sci. Eng. A 529 (2011) 177–183. [10] Q.M. Gong, Z. Li, X.D. Bai, D. Li, J. Liang, Compos. Sci. Technol. 65 (2005) 1112–1119. [11] M. Guellali, R. Oberacker, M.J. Hoffmann, Carbon 43 (2005) 1954–1960. [12] R. Naslain, Composites A 29 (1998) 1145–1155. [13] H.L. Li, H.J. Li, J.H. Lu, C. Sun, Y.J. Wang, D.J. Yao, et al., Mater. Sci. Eng. A 530 (2011) 57–62.
Contents lists available at SciVerse ScienceDirect
Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea
Rapid communication
The reinforcement and toughening of pyrocarbon-based carbon/carbon composite by controlling carbon nanotube growth position in carbon felt Qiang Song a, Ke-zhi Li a,n, Le-hua Qi b, He-jun Li a, Jin-hua Lu a, Lei-lei Zhang a, Qian-gang Fu a a b
State Key Laboratory of Solidification Processing, Northwestern Polytechnical University, Xi’an 710072, China School of Mechanical Engineering, Northwestern Polytechnical University, Xi’an 710072, China
a r t i c l e i n f o
abstract
Article history: Received 18 July 2012 Received in revised form 19 November 2012 Accepted 21 November 2012 Available online 29 November 2012
The flexural properties of pyrocarbon-based carbon/carbon (C/C) composites reinforced by in situ grown carbon nanotubes are reported. The effect of nanotube growth position (on carbon fibers or in the spaces between them) was investigated. A method of reinforcing and toughening C/C composites is proposed and the corresponding mechanisms are discussed. Crown Copyright & 2012 Published by Elsevier B.V. All rights reserved.
Keywords: Mechanical characterization Composites Fracture
Fig. 1. SEM images of CNTs with different growth positions: (a) CNTs grown on carbon fibers and (b) CNTs grown in the spaces between fibers.
1. Introduction Over the past few decades, much different effort has been devoted to improve the toughness of carbon/carbon (C/C) composites [1–3]. An efficient work is adding nano-filler, such as carbon nanofiber (CNF), into the composites [2,3]. In the cases, resin or pitch carbon was used as the matrix and the toughening mechanisms have been found to be ‘‘CNF bridging’’ and ‘‘CNF pull-out’’. However, for C/C
n
Corresponding author. Fax: þ86 29 8849 5764. E-mail address: [email protected] (K.-z. Li).
composites with pyrocarbon (PyC) matrix, the existing reports show that adding nano-fillers such as carbon nanotube (CNT) always leads to the brittle fracture of the composites despite a largely increased mechanical strength [4–6]. The reasons for the brittle fracture are various.One of them is the failure of the toughening mechanisms mentioned above because the bonding between CNT and PyC is very strong [7] that inhibits the long-distance pull-out of CNT from carbon matrix. Another important one is the overly-enhanced F/M interface bonding because of the introduction of CNTs onto carbon fibers which blocks the fiber pull-out and decreases the fracture toughness [5,6], even the CNT content is very low [5].
0921-5093/$ - see front matter Crown Copyright & 2012 Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.msea.2012.11.074
72
Q. Song et al. / Materials Science & Engineering A 564 (2013) 71–75
Recently, it is found that carbon matrix with multilayer microstructures is beneficial to the mechanical properties of C/C composites, especially the toughness [8,9]. On the other hand, introducing
CNTs into the matrix area of C/C composites can induce the formation of PyC matrix with multilayer microstructures during chemical vapor deposition (CVD) [9,10], under the precondition of not increasing the F/M interface bonding. Thus, it is very possible to develop a novel method of reinforcing and meanwhile toughening PyC-based C/C composites by adding CNTs into the space between carbon fibers rather than onto fibers. In the present work, this issue was demonstrated based on the investigation into the effects of CNT growth position (on the fibers or in the spaces between them) on the flexural properties of C/C composites.
2. Experimental
Fig. 2. Representative flexural stress–strain curves of the three composites.
Table 1 Mechanical properties of the three kinds of C/C composites. Composites CNT content (wt.%)
CNT Bulk length(mm) density (g m-3)
Flexure strength (MPa)
FD
C/C CNTP1-C/C
0 2–4 4–20 8–20 5–18 5–20
143 75 182 78 251 77 179 710 196 712 212 78
0.27 7 0.07 0.037 0.01 o0.01 0.56 7 0.10 0.58 7 0.08 0.49 7 0.10
CNTP2-C/C
0 2.0 9.5 1.9 6.2 10.3
1.55 1.57 1.56 1.57 1.57 1.56
Carbon felts (bulk density: 0.43 g/cm3; fibers diameter: 79 mm) were used as preform materials. The growth of CNTs in carbon felts was accomplished by catalytic CVD using Ni(NO3)2 6H2O as catalyst precursor. Incipient wetness technique was adopted to introduce catalyst into felts using acetone as solvent. The gas system for CNT growth was CH4/N2 (flow ratio: 1/10). The growth temperature was 1020 1C. The growth time was 1 h. To obtain CNTs grown on carbon fibers (named CNTs-P1), before dipped in catalyst solution, carbon felts were firstly coated with a layer of PyC (several dozen nanometers) to prevent the dissolution of metal catalyst into carbon fibers and then were treated in 65 vol.% HNO3 at 60 1C for 10 h. To obtain CNTs grown in the spaces between fibers (named CNTs-P2), carbon felts were directly dipped in catalyst solution. After CNT growth, all the felts were densified in a CVD reactor under rough laminar (RL) PyC deposition conditions [11]. For comparison, C/C composites with the same felts were prepared under the same CVD conditions. The flexural properties of the three composites were tested by three-point-bending test at room temperature, conducted at loading speed of 0.5 mm/min and support span of 20 mm length.
Texture of RL PyC
Fig. 3. SEM microstructure and its model of the three composites: (a) C/C composites; (b) CNTP1-C/C composites; (c) CNTP2-C/C composites (the inset shows that RL PyC is deposited nearly around a single CNT).
Q. Song et al. / Materials Science & Engineering A 564 (2013) 71–75
The size of bending specimens was 35 mm 5 mm 2 mm. The number of specimens used in the flexural test was no less than five for every test.
3. Results and discussion Scanning electron microscopy (SEM) image (Fig. 1a) shows CNTs-P1, with straight body and length of 2 4 mm, have
73
approximately radial growth orientation (relative to carbon fibers), forming a dense three dimensional (3D) network structure (Fig. 1a inset). CNTs-P2 (Fig. 1b) have long length (5 18 mm) and a scattered distribution in the spaces between fibers. CNTs-P1 and CNTs-P2 account for 2.0 wt.% and 1.9 wt.% of the hybrid felt, respectively. Fig. 2 shows the representative flexural stress–strain curves of the three composites. After adding CNTs-P1 and CNTs-P2, similar mechanical strength improvements are obtained and the flexural
Fig. 4. SEM images of fracture surfaces for C/C composites (a, d), CNTP1-C/C composites (b, e), CNTP2-C/C composites (c, f).
74
Q. Song et al. / Materials Science & Engineering A 564 (2013) 71–75
Fig. 5. Enlarged SEM images of the fractured CNTP2-C/C composites: (a) shows the rough surface of fractured PyC matrix, in which interfaces between CNT/PyC units and fractured CNT/PyC structures can be seen; (b) shows a fractured CNT/PyC structure with step-like shape and some delaminated RL PyC sheets around it.
strength increases by 27% and 25%, respectively (Table 1). However, the fracture behaviors of the two composites are greatly different. The composites reinforced by CNTs-P2 (named CNTP2C/C composites) and C/C composites exhibit pseudo-plastic fracture behaviors [11]; while, C/C composites reinforced by CNTs-P1 (CNTP1-C/C composite) shows a typically brittle fracture. Quantitative comparison is done based on the calculation of ductility factor FD (which can estimate the quasi-ductile fracture behavior of C/C composites [11]). As shown in Table 1, CNTP2-C/C composites have over 1700% and 107% FD improvements, respectively compared with CNTP1-C/C and C/C composites. The great differences between the fracture behaviors of the two CNT-hybridized composites can be explained by their greatly different microstructures. Without CNTs, RL PyC is deposited layer by layer around carbon fiber (Fig. 3a). After adding CNTsP1, PyC with very random orientation (i.e. isotropic (ISO) PyC) is deposited nearly around fibers (Fig. 3b) though the deposition conditions should lead to RL PyC. Only in the area away from CNTs, layered PyC is formed. The formation of ISO PyC is related to the very disorderly internal structure of the 3D CNT network [5]. ISO structures, which are seriously intertwisted, can immensely inhibit the deflection of matrix microcracks in F/M interface area and then lead to the brittle fracture of CNTP1-C/C composites [12] though can increase the mechanical strength obviously [9]. More importantly, the strongly enhanced F/M interface bonding due to the radially grown CNTs (Fig. 4e inset, CNT pull-out can be seen), characterized by the absent fiber pullout (Fig. 4b and e), further leads to the high strength and the poor ductility of CNTP1-C/C composites. After adding CNTs-P2, RL PyC is deposited around CNT (Fig. 3c inset) rather than fiber in matrix area and CNT/RL unit is formed; while, in F/M interface area, RL PyC is still deposited around carbon fibers (labeled by dotted arrows in Fig. 3c). Due to the similar microstructure of F/M interface, the fiber pull-out situation of fractured CNTP2-C/C composites is similar to that of fractured C/C composites (Fig. 4c, f and 4a, d). Compared with C/C composites, the obviously increased mechanical strength and largely improved toughness of CNTP2-C/C composites should be attributed to multilayer carbon matrix. Concretely, CNT/RL units and the interfaces between them constitute multilayer structures, which lead destructive cracks to spread along multiple paths [8] (Fig. 4f shows a rough fracture surface of PyC matrix of CNTP2-C/C composites; while, the fracture surface of C/C composites shows very flat, Fig. 4d inset). More energies are dissipated during this course, which can increase mechanical strength and also make flexural stress release gently [13], leading to sliding regions occurred in stressstrain curve which corresponds to an increased FD. The multiple
paths mentioned above mainly include the interfaces between CNT/RL units (Fig. 5a) and CNT/RL unit itself (Fig. 5b shows the step-like fracture of a CNT/PyC unit and the delaminated RL PyC sheets around it). In addition, CNT pull-out has also been found (Fig. 5b); however, the quantity of pulled-out CNTs is very small and the length is also very short (100–500 nm). In contrast, for CNT-reinforced pitch-based C/C composites, the quantity of pulled-out CNTs is considerable and the pull-out length ranges from 3 mm to over 5 mm [3]. Thus, it can be said that the contribution of CNT pull-out to the high strength and the high ductility must be very small. In our work, it is also found that increasing the mass content of CNTs-P2 can further reinforce and toughen C/C composites, as shown in Table 1. When this content is over 10 wt.%, the ductility of the composites will decrease, however, the FD value is still much larger than that of CNTP1-C/C and C/C composites. On the other hand, increasing the mass content of CNTs-P1 can further lower the ductility, though which can also endow a very high flexural strength to C/C composites. Anyway, the significance of these results consists in the fact that adding CNTs into matrix is indeed an effective method of reinforcing and also toughening C/C composites.
4. Conclusions In this work, the effects of CNT growth position on the flexural properties of PyC-based C/C composites were investigated. Results show that adding CNTs in the spaces between carbon fibers can not only enhance the mechanical strength but also drastically improve the fracture toughness of C/C composites, compared with adding CNTs on fibers. Different from the previous work [2,3], the excellent mechanical properties benefit from the multilayer structured PyC matrix formed under the influence of CNTs dispersed in the spaces between fibers rather than the roles of CNT bridging and CNT pull-out during the fracture of C/C composites.
Acknowledgments This work has been supported by the program of introducing talents of discipline to universities (Grant no. B08040) and the Ph.D. thesis innovation fund of NWPU (cx201214), key grant project of Chinese Ministry of Education (313047) and National Natural Science Foundations of China (51275417, 50832004 and 51202194).
Q. Song et al. / Materials Science & Engineering A 564 (2013) 71–75
References [1] G. Chollon, O. Siron, J. Takahashi, H. Yamauchi, K. Maeda, K. Kosaka, Carbon 39 (2001) 2065–2075. [2] H.L. Liu, Z.H. Jin, Z.B. Hao, X.M. Zeng, Mater. Sci. Eng. A 483–484 (2008) 316–318. [3] H.L. Li, H.J. Li, J.H. Lu, K.Z. Li, C. Sun, D.S. Zhang, Mater. Sci. Eng. A 547 (2012) 138–141. [4] P. Xiao, X.F. Lu, Y.Q. Liu, L.L. He, Mater. Sci. Eng. A 528 (2011) 319–323. [5] K.Z. Li, Q. Song, L.Y. Wu, H.J. Li, Q.J. Gong, J. Inorg. Chem. 27 (2011) 1001–1007. [6] Q. Song, K.Z. Li, H.L. Li, H.J. Li, C. Ren, Carbon 50 (2012) 3949–3952. [7] H. Allouche, M. Monthioux, Carbon 43 (2005) 1265–1278.
75
[8] H.J. Li, H.L. Li, K.Z. Li, Y.J. Wang, D.S. Zhang, J.H. Lu, J. Mater. Sci. 46 (2011) 4667–4674. [9] K.Z. Li, H.L. Deng, H.J. Li, X. Li, X.N. Zhao, H.J. Luo, Mater. Sci. Eng. A 529 (2011) 177–183. [10] Q.M. Gong, Z. Li, X.D. Bai, D. Li, J. Liang, Compos. Sci. Technol. 65 (2005) 1112–1119. [11] M. Guellali, R. Oberacker, M.J. Hoffmann, Carbon 43 (2005) 1954–1960. [12] R. Naslain, Composites A 29 (1998) 1145–1155. [13] H.L. Li, H.J. Li, J.H. Lu, C. Sun, Y.J. Wang, D.J. Yao, et al., Mater. Sci. Eng. A 530 (2011) 57–62.