- Email: [email protected]
Internal friction behaviour of carbon fibre reinforced multilayered (PyC–SiC)n matrix composites
Accepted Manuscript Internal friction behaviour of carbon fibre reinforced multilayered (PyC–SiC)n matrix composites Yan Jia, Kezhi Li, Lizhen Xue, Junjie Ren, Shouyang Zhang PII:
S1359-8368(16)32576-8
DOI:
10.1016/j.compositesb.2017.01.068
Reference:
JCOMB 4879
To appear in:
Composites Part B
Received Date: 4 November 2016 Revised Date:
9 December 2016
Accepted Date: 28 January 2017
Please cite this article as: Jia Y, Li K, Xue L, Ren J, Zhang S, Internal friction behaviour of carbon fibre reinforced multilayered (PyC–SiC)n matrix composites, Composites Part B (2017), doi: 10.1016/ j.compositesb.2017.01.068. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Internal friction behaviour of carbon fibre reinforced multilayered (PyC–SiC)n matrix composites
RI PT
Yan Jia, Kezhi Li∗, Lizhen Xue, Junjie Ren, Shouyang Zhang† State Key Laboratory of Solidification Processing, Carbon/carbon composites Research Centre, Northwestern Polytechnical University, Xi’an 710072, PR China
SC
Abstract
Carbon fibre reinforced multilayered (PyC–SiC)n matrix (C/(PyC–SiC)n)
M AN U
composites were prepared by alternate deposition of pyrocarbon (PyC) and SiC inside preforms via chemical vapour infiltration. The matrix microstructures and internal friction behaviours of C/(PyC–SiC)n composites (n=1, 2 and 4) under different testing conditions of frequency, strain amplitude and temperature were studied. The results
TE D
show that with increasing the number of sequences (n), the internal friction increases due to the enhancement of interfacial internal friction. The internal friction of C/(PyC– SiC)n composites increases with the increase of frequency related to thermoelastic
EP
mechanism, but exhibits anomalous amplitude effect in the testing amplitude range. Effect of temperature on internal friction behaviours is attributed to combined effects of
AC C
carbon fibres, PyC–SiC matrices and interfaces between fibres and PyC, adjoining PyC and SiC layers. It is also found that internal friction is sensitive to microstructural defects induced by damage. These results indicate that internal friction is an effective and efficient method to characterize the structural evolution and internal damage of C/(PyC–SiC)n composites non-destructively. * Corresponding author. Tel.: +86 29 88495764; E-mail address: [email protected] (K. Li) † Corresponding author. Tel.: +86 29 88492272; E-mail address: [email protected] (S. Zhang) 1
ACCEPTED MANUSCRIPT
Keywords: A. Carbon–carbon composites (CCCs); A. Layered structures; B. Internal friction/damping; E. Chemical vapour deposition (CVD)
RI PT
1. Introduction Internal friction or damping capacity of a material refers to its ability to convert mechanical vibration energy to internal energy irreversibly, which is an important
SC
parameter in mechanical design [1]. Internal friction depends on the constitutions and
microstructure of materials and is very sensitive to the variation of microstructure [2].
M AN U
Importantly, it is a non-destructive detection method as an effective and efficient tool to study microstructural evolution within materials [3]. Thus, internal friction not only plays an enabling role in evaluating dynamic mechanical properties, but also is a basic approach for characterizing structural changes. Up to now, enormous work has been
TE D
carried out to internal friction analysis for studying composites, including metal-based [4-8], polymer-based [9-12], carbon-based [13-18] and SiC-based composites [19-23]. Various mechanisms and models have been established to predict the internal friction of
EP
composite materials [24-27].
Carbon fibre reinforced carbon–silicon carbide matrix (C/C–SiC) composites have
AC C
been widely used as one of the most promising structural materials due to their superior properties, such as low density, high strength at elevated temperature and excellent friction performance [28-30]. C/C–SiC composites are also potential candidates for brake discs and clutch systems [31], which involve structural vibration and internal friction. Therefore, it is necessary to study the behaviour of internal friction of C/C–SiC composites. 2
ACCEPTED MANUSCRIPT
Currently, many preparation processes have been developed to manufacture C/C– SiC composites, including chemical vapour infiltration (CVI) [32], polymer
RI PT
impregnation pyrolysis (PIP) [33, 34] and liquid silicon infiltration (LSI) [35]. It is well known that PIP process would inevitably produce many defects resulting from gas
volatilization and shrinkage within matrix during pyrolysis process and LSI process
SC
requires high temperature (above 1410 °C) for reaction of C and Si leading to great
residual stress and matrix cracking [36], which can severely deteriorate the performance
M AN U
of C/C–SiC composites. Inversely, C/C–SiC composites prepared by CVI have a good matrix and controlled microstructure [36], indicating that CVI is an effective way to obtain C/C–SiC composites with high performance.
In our previous work [37], a new member of C/C–SiC composites was fabricated
TE D
by alternate deposition of pyrocarbon (PyC) and SiC inside preforms via isothermal CVI (ICVI), i.e. carbon fibre reinforced multilayered (PyC–SiC)n matrix (C/(PyC–SiC)n) composites. Static mechanical and electromagnetic interference shielding properties of
EP
the composites were studied intensively. However, the fundamental internal friction of C/(PyC–SiC)n composites has yet to be reported. In this paper, the internal friction
AC C
behaviours of C/(PyC–SiC)n composites with various microstructures were investigated.
The internal friction mechanisms under different testing conditions were also discussed in detail.
2. Experimental procedure 2.1 Materials preparation 2.5D needle-punched carbon fibre felts (Yixing Tianniao High Technology Co. Ltd., 3
ACCEPTED MANUSCRIPT
China) were used as preforms. The fibre type was commercial 12 k T700 PAN-based (Toray, Japan) carbon fibre with a filament diameter of 7 µm, and fibre volume fraction
RI PT
of the preform was about 25%. Schematic diagram of preform architecture is shown in Fig. 1(a). It was fabricated by repeatedly overlapping layers of 0° non-woven carbon fibre cloth, short-cut fibre web, 90° non-woven carbon fibre cloth and short-cut fibre
SC
web with needle-punching step by step. ICVI was employed to alternately infiltrate PyC and SiC into preforms. One PyC–SiC sequence ((PyC–SiC)1), two PyC–SiC sequences
M AN U
((PyC–SiC)2) and four PyC–SiC sequences ((PyC–SiC)4) were deposited on carbon fibres to prepare three kinds of C/(PyC–SiC)n composites (designated as CS1, CS2 and CS4, respectively), as illustrated in Fig. 1(b). PyC matrix was deposited at 1100 °C under a negative pressure of 5 kPa with the flow rate of methane at 1.2 L/min. As for
TE D
the deposition process of SiC matrix, methyltrichlorosilane (MTS, CH3SiCl3) was used as precursor and carried by bubbling hydrogen in gas phase, and argon was employed as dilute gas. The deposition temperature was controlled at 1100 °C under a pressure of 3
EP
kPa, molar ratio of H2 to MTS was 10, and flow rate of Ar was 300 mL/min. Further details of deposition time for each layer of PyC and SiC were described elsewhere [37].
AC C
The final bulk density and open porosity of each composite were 1.34 g/cm3 and 35.1%,
1.33 g/cm3 and 37.2%, 1.36 g/cm3 and 34.6%, respectively. 2.2 Characterization
The internal friction tests were performed on a dynamic mechanical analyser (DMA, Q800, TA, USA) by mode of three-point bending force vibration. Samples were cut into rectangular bars with dimensions of 40 mm × 5 mm × 3 mm. The loading 4
ACCEPTED MANUSCRIPT
direction was perpendicular to the non-woven cloth. The testing conditions were given as follows: frequency was from 0.1 to 10 Hz at strain amplitude of 9 µm and room
RI PT
temperature (RT), strain amplitude varied from 5 to 10 µm at frequency of 5 Hz and RT, temperature range was from RT to 300 °C with a heating rate of 5 °C/min at frequency of 5 Hz and strain amplitude of 9 µm. All runs were conducted in air. The internal
SC
friction (Q-1) was calculated according to the following equation [38]: Q-1 = tan δ = E''/E'
(1)
M AN U
where δ is the loss angle between applied stress and strain, E' is the dynamic storage (real) modulus and E'' is the dynamic loss (imaginary) modulus.
Microstructures of C/(PyC–SiC)n composites were observed under polarized light microscope (PLM, DMLP, Leica, Germany). Polished cross-sectional morphology of
TE D
the composites was characterized by scanning electron microscope (SEM, VEGA3, Tescan, Czech) equipped with energy dispersive spectroscope (EDS, Oxford INCA, UK).
EP
3. Results
3.1 Microstructure characterization
AC C
Fig. 2 shows the backscattered electron (BSE) images of polished cross-sectional
morphology and EDS mapping of insets of C/(PyC–SiC)n composites (n=1, 2 and 4). It
can be seen that the matrices of C/(PyC–SiC)n composites are composed of binary phases displaying dark grey and light grey. Additionally, the matrices become increasingly fine and close in texture with the increase of PyC–SiC sequences number (n). Under short deposition time of 40 h, there remain a large number of inter-bundle 5
ACCEPTED MANUSCRIPT
and intra-bundle pores. Enlarged views of dotted boxes (insets of Fig. 2(a)-(c)) display that the dark grey phase is carbon and light grey phase is SiC, attesting by
RI PT
corresponding EDS mapping of insets. It can be clearly observed from the insets that CS1, CS2 and CS4 have a two-layer, four-layer and eight-layer structure around the carbon fibres. Owing to the effective ICVI process, PyC and SiC are sequentially
SC
well-infiltrated into the preforms and formed multilayered matrices as designed.
Fig. 3 presents the PLM micrographs of CS1, CS2 and CS4, which show various
M AN U
types of matrix microstructures. The multilayered matrices exhibit alternately black and white concentric annuluses marked by arrows, which could be attributed to different light reflectivity under PLM. PyC has high optical activity displaying white, whereas SiC is insensitive to the polarized light appearing black. It is worth noting that PLM
TE D
images of layered microstructures consist with BSE images (insets of Fig. 2) apart from different indicative colours of PyC and SiC. 3.2 Internal friction
EP
3.2.1 Internal friction versus frequency
Internal friction of C/(PyC–SiC)n composites as a function of frequency at strain
AC C
amplitude of 9 µm and RT is illustrated in Fig. 4. Among the three composites, CS4 has
the highest internal friction and CS1 has the lowest internal friction in the whole testing frequency. The internal friction of all composites is almost unchanged from 0.1 to 2 Hz, and then begins to increase until 10 Hz. The similar relationship between internal friction and frequency is also reported by Hou et al. [14] and Cheng et al. [17] in C/C composites. The results indicate that high frequency could affect the internal friction 6
ACCEPTED MANUSCRIPT
behaviours of C/(PyC–SiC)n composites. 3.2.2 Internal friction versus strain amplitude
RI PT
Fig. 5 displays internal friction of C/(PyC–SiC)n composites versus strain amplitude at frequency of 5 Hz and RT. It is obvious that the internal friction is
dependent on testing amplitude. As the strain amplitude increases from 5 to 10 µm, the
SC
internal friction of all composites decreases gradually. It is also observed that the overall level of internal friction in CS4 is higher than that in CS1 and CS2.
M AN U
3.2.3 Internal friction versus temperature
Fig. 6 shows the dependence of internal friction on temperature at frequency of 5 Hz and strain amplitude of 9 µm for C/(PyC–SiC)n composites. The internal friction values are in the sequences of CS4 > CS2 > CS1 over the whole testing temperature
TE D
range. Moreover, all internal friction-temperature curves can be divided into three stages with the increase of temperature. The internal friction declines monotonically from RT to about 50 °C in stage I. In stage II, the internal friction curves of CS1 and
EP
CS2 keeps almost constant from 50 to 200 °C in spite of gentle fluctuations, while the
AC C
curve of CS4 fluctuates greatly. When temperature is above 200 °C in stage III, the internal friction of all composites decreases again with varying slopes. Especially, two internal friction peaks appear at about 75 and 240 °C for CS4 in stage II and III, respectively. The similar phenomena are also found in C/C [39], SiC/SiC [19] and C/SiC [21] composites except for the different peak temperatures. 3.2.4 Internal friction versus frequency before and after damage Fig. 7 presents the internal friction of CS2 versus frequency before and after slight 7
ACCEPTED MANUSCRIPT
damage, which is conducted by a mild three-point bending. The obtained samples still have highly structural integrity and cannot be distinguished from other normal samples
RI PT
through appearance. As shown in Fig. 7, the internal friction after slight damage is much higher than that before damage, which demonstrates that the internal friction is very sensitive to damage.
SC
4. Discussion 4.1 Intrinsic internal friction
M AN U
It has been recognized that the internal friction behaviour of a composite is determined not only by matrix and fibres, but also by properties of interfaces within the composite [40]. Internal friction (Q-1) is the combined results of such various factors, which can be described as follows:
(2)
TE D
Q-1 = tan δf + tan δm + tan δint
where tan δf, tan δm and tan δint are the internal friction caused by fibres, matrix and interfaces, respectively.
EP
For all three composites, they have the same carbon felt preforms, similar density, open porosity, PyC and SiC content [37]. In addition, the interfaces between carbon
AC C
fibres and PyC are determined by the volume fraction of fibres in C/(PyC–SiC)n
composites [15]. Thus, the contributions of carbon fibres, PyC-SiC matrix and fibre/PyC interfaces to internal friction are approximately equal among them. The difference of internal friction can be attributed to interfaces between PyC and SiC (PyC/SiC) within the matrices. Eq. (2) can be represented as follows: Q-1 = tan δf + tan δm + tan δint, 1 + tan δint, 2 = tan δs + tan δint, 2 8
(3)
ACCEPTED MANUSCRIPT
where tan δint, 1 and tan δint, 2 are internal friction resulting from fibre/PyC and PyC/SiC interfaces, respectively. tan δs is the sum of tan δf, tan δm and tan δint, 1.
RI PT
In general, weak or strong interfacial bonding both can contribute to the internal friction. Many researchers have found that interfacial internal friction (tan δint) increases with increasing fibres or particulates volume fraction [4, 14, 15]. Obviously, as volume
SC
fraction of reinforcement increases, the number of reinforcement/matrix interfaces
increases. As shown in Fig. 2 and 3, the number of PyC/SiC interfaces increases with
M AN U
the increase of sequences number (n), leading to tan δint, 2 (CS4) > tan δint, 2 (CS2) > tan δint, 2 (CS1). Combined with previous analysis, it can be concluded that Q-1 (CS4) > Q-1 (CS2) > Q-1 (CS1) regardless of external influences. This could explain the growing trend of internal friction in Fig. 4-6 nicely.
TE D
4.2 Effects of frequency and amplitude on internal friction
In the experiments, the testing samples are forcedly vibrated under flexural stress, which could generate cyclic heat flow from the region of compressive stress to the
EP
region of tensile stress in the inhomogeneous composites (Fig. 8) [25, 26]. The temperature gradients are produced resulting in dissipation of mechanical energy.
AC C
Therefore, the thermoelastic mechanism should be given priority as the most fundamental internal friction mechanisms [1, 25]. According to the approximate relationship between frequency and internal friction proposed by Zener [41], thermoelastic internal friction would increase with increasing frequency and could be ignored at low frequency, which is in a good agreement with the result in Fig. 4. Moreover, the thermoelastic internal friction is frequency-dependent and 9
ACCEPTED MANUSCRIPT
amplitude-independent [39]. However, as shown in Fig. 4 and 5, the internal friction of C/(PyC–SiC)n composites is influenced by not only frequency but also amplitude. It is
RI PT
reasonable to suppose that there exist other mechanisms affecting the internal friction behaviours of C/(PyC–SiC)n composites.
Theoretically, the amplitude dependence is usually relevant to static hysteresis [38].
SC
As opposed to thermoelastic mechanism, it is independent of frequency and dependent on amplitude [1]. During measurement, the stress-strain curves of C/(PyC–SiC)n
M AN U
composites will give hysteresis loop under cyclic stress, which can be explained by the dislocation mechanism (K-G-L theory) proposed by Granato and Lücke [42]. The area enclosed by loop corresponds to the dissipated energy. In such circumstance, the internal friction would increase with the increase of strain amplitude. But in the present
TE D
work, the internal friction of all three composites reduces gradually as amplitude increases (Fig. 5), presenting anomalous amplitude effect [43]. The results demonstrate that other mechanisms are coupled with static hysteresis mechanism in terms of
EP
amplitude-dependence, which remains to be furtherly studied. 4.3 Effect of temperature on internal friction
AC C
The variation of internal friction-temperature curves becomes complicated
compared with internal friction-frequency and internal friction-strain amplitude curves, which could be attributed to the specific characteristics of components (fibre, matrix and interface) versus temperature. The internal friction of C/C composites and graphite has been found to decrease with increasing temperature [16, 39]. On the contrary, the internal friction of CVI SiC/SiC composites and SiC increase as temperature increases 10
ACCEPTED MANUSCRIPT
[19, 44]. It can be concluded that the different variation trends are intrinsic properties of carbon and SiC materials. Furthermore, the thermal stress and interfacial delamination
RI PT
are induced due to the mismatches of coefficient of thermal expansion (CTE) between carbon fibres and PyC, PyC and SiC during the cooling stage [45]. Hence, the overall
internal friction behaviours can be understood by taking into account all factors, which
SC
may compete with each other.
In low temperature range (stage I), the carbon fibres and PyC–SiC matrix are
M AN U
predominant and the internal friction is controlled by the behaviours of carbon and SiC. The carbon content is higher than SiC content [37] and the internal friction of carbon (5~8 × 10-3 [22]) is also larger than that of SiC (1.2~5 × 10-3 [22]). Thus, carbon material plays a dominant role, resulting in the decline of internal friction. When
TE D
temperature rises, the thermal stress at fibre/matrix and PyC/SiC interfaces can be released and delamination may be reduced or sealed gradually, which will improve the bonding of fibre/matrix interfaces and adjoining PyC and SiC layers. The attrition
EP
energy loss increases leading to the enhancement of internal friction. On the other hand, the character of SiC material also contributes to the increase of internal friction in high
AC C
temperature. The combined positive contributions could compensate the negative contribution of carbon material character, so the internal friction of CS1 and CS2 remains nearly stable in stage II. As temperature continues to rise (stage III), the internal friction produced by different kinds of defects including dislocation, debonding and microcracks would decrease due to the decrement of internal friction sources according to the dislocation mechanism [42]. Therefore, the overall internal friction decreases. 11
ACCEPTED MANUSCRIPT
Compared with CS1 and CS2, the occurrence of two peaks in CS4 could be the competitive results between a greater deal of interfacial internal friction and character of
RI PT
carbon itself. 4.4 Effect of damage on internal friction
Fig. 9 shows the SEM image of defects in CS2 after slight damage. Though CS2
SC
endures slight damage, matrix microcracks and debonding between fibres and PyC are
generated, which is relevant to the significant differences of internal friction before and
M AN U
after damage (Fig. 7). Under cyclic stress, such defects could dissipate much energy resulting in sharp increase of internal friction. Therefore, internal friction is a good indicator to evaluate the occurrence of damage, especially micro-damage. 5. Conclusions
TE D
In the present work, internal friction behaviours of C/(PyC–SiC)n composites with different number of PyC–SiC sequences (n=1, 2 and 4) were studied under testing conditions of frequency, strain amplitude and temperature, respectively. The sequences
EP
number (n) exerts an important influence on the internal friction of C/(PyC–SiC)n composites. With increasing the sequences number (n), the interfacial internal friction
AC C
increases resulting in Q-1 (CS4) > Q-1 (CS2) > Q-1 (CS1) under different testing conditions. The internal friction keeps unchanged at lower frequency and increases at higher frequency range, which can be explained by thermoelastic mechanism. The internal friction of all composites decreases in the whole testing strain amplitude, but the mechanisms of anomalous amplitude effect are still not clear. The internal friction-temperature dependence is controlled by interfacial characteristic and intrinsic 12
ACCEPTED MANUSCRIPT
properties of carbon and SiC, exhibiting different trends in three stages. Moreover, the superior responsiveness of internal friction to damage demonstrates that it is a
RI PT
non-destructive detection method to determine the internal damage of materials effectively and efficiently. Acknowledgements
SC
This work was supported by the National Natural Science Foundation of China (Grant Nos. 51521061 and 51472202) and “111” Project (Grant No.B08040).
1.
M AN U
References
Blanter MS, Golovin I, Neuhauser H, Sinning H. Internal friction in metallic
materials. A handbook Series: Springer Series in Materials Science, 2007. 2.
Berthelot J-M, Assarar M, Sefrani Y, Mahi AE. Damping analysis of composite
3.
TE D
materials and structures. Compos Struct 2008;85(3):189-204.
Boccaccini AR, Ponton CB, Chawla KK. Development and healing of matrix
microcracks in fibre reinforced glass matrix composites: assessment by internal friction.
4.
EP
Mater Sci Eng A 1998;241(1–2):141-50.
Perez RJ, Zhang J, Gungor MN, Lavernia EJ. Damping behavior of 6061Al/Gr
AC C
metal matrix composites. Metall Trans A 1993;24(3):701-12. 5.
Parrini L, Schaller R. Thermal stresses in metal matrix composites studied by
internal friction. Acta Mater 1996;44(12):4881-8. 6.
Zhang HX, Gu MY. Internal friction behavior in SiC particulate reinforced
aluminum metal matrix composite in thermal cycling. J Alloy Compd 2006;426(1– 2):247-52. 13
ACCEPTED MANUSCRIPT
7.
Hu J, Wu GH, Zhang Q, Gou HS. Mechanical properties and damping capacity of
SiCp/TiNif/Al composite with different volume fraction of SiC particle. Compos Part B:
8.
RI PT
Eng 2014;66:400-6. Shin J, Choi K, Shiko S, Choi H, Bae D. Mechanical damping behavior of
Al/C60-fullerene composites with supersaturated Al–C phases. Compos Part B: Eng
9.
SC
2015;77:194-8.
Feng JM, Guo ZS. Effects of temperature and frequency on dynamic mechanical
M AN U
properties of glass/epoxy composites. J Mater Sci 2016;51(5):2747-58.
10. Babal AS, Singh BP, Jyoti J, Sharma S, Arya AK, Dhakate SR. Synergistic effect on static and dynamic mechanical properties of carbon fiber-multiwalled carbon nanotube hybrid polycarbonate composites. RSC Adv 2016;6(72):67954-67.
TE D
11. Rafiee F, Otadi M, Goodarzi V, Khonakdar HA, Jafari SH, Mardani E, et al. Thermal and dynamic mechanical properties of PP/EVA nanocomposites containing organo-modified layered double hydroxides. Compos Part B: Eng 2016;103:122-30.
EP
12. Ashworth S, Rongong J, Wilson P, Meredith J. Mechanical and damping properties of resin transfer moulded jute-carbon hybrid composites. Compos Part B: Eng
AC C
2016;105:60-6.
13. Qiao SR, Zhou SR, Bai SH, Sun XY. The internal friction of unidirectional CC composites fabricated in a magnetic field. Carbon 1997;35(3):389-92. 14. Hou XH, Li HJ, Shen J, Wang C, Zhu ZG. Effects of microstructure on the internal friction of carbon–carbon composites. Mater Sci Eng A 2000;286(2):250-6. 15. Wang C, Zhu ZG, Hou XH, Li HJ. Damping characteristics of CVI-densified 14
ACCEPTED MANUSCRIPT
carbon–carbon composites. Carbon 2000;38(13):1821-4. 16. Kogo Y, Iijima Y, Igata N, Ota Ki. Internal friction of carbon–carbon composites at
RI PT
elevated temperatures. J Alloy Compd 2003;355(1–2):148-53. 17. Cheng J, Li HJ, Zhang SY, Xue LZ, Luo WF. Internal friction behavior of
unidirectional carbon/carbon composites after different fatigue cycles. Mater Sci Eng A
SC
2014;600:129-34.
18. Liu Y, Fu QG, Zhao FL, Sun GD, Li HJ. Internal friction vs. thermal shock in C/C
M AN U
composites. Compos Part B: Eng 2016;106:59-65.
19. Sato S, Serizawa H, Araki H, Noda T, Kohyama A. Temperature dependence of internal friction and elastic modulus of SiC/SiC composites. J Alloy Compd 2003;355(1–2):142-7.
TE D
20. Zhang J, Xu YD, Zhang LT, Cheng LF. Internal friction and dynamic modulus of three-dimensional silicon carbide-matrix composites. Mater Lett 2005;59(19– 20):2535-8.
EP
21. Zhang Q, Cheng LF, Wang W, Zhang LT, Xu YD. Effect of SiC coating and heat treatment on damping behavior of C/SiC composites. Mater Sci Eng A 2008;473(1–
AC C
2):254-8.
22. Hou JT, Qiao SR, Lu GF, Zhang CY, Zhang YB. Influence of heat treatment on the internal
friction
of
2D-C/SiC
composites.
J
Mater
Process
Technol.
2009;209(7):3555-60. 23. Hong ZL, Cheng LF, Zhang LT, Wang YG. Internal Friction Behavior of C/SiC Composites with Environmental Barrier Coatings in Corrosive Environment. Int J Appl 15
ACCEPTED MANUSCRIPT
Ceram Tec 2011;8(2):342-50. 24. Chandra R, Singh SP, Gupta K. Micromechanical damping models for
RI PT
fiber-reinforced composites: a comparative study. Compos Part A: Appl Sci Manuf 2002;33(6):787-96.
25. Chandra R, Singh SP, Gupta K. Damping studies in fiber-reinforced composites-a
SC
review. Compos Struct 1999;46(1):41-51.
26. Chandra R, Singh SP, Gupta K. A study of damping in fiber-reinforced composites.
M AN U
J Sound Vib 2003;262(3):475-96.
27. Treviso A, Van Genechten B, Mundo D, Tournour M. Damping in composite materials: Properties and models. Compos Part B: Eng 2015;78:144-52. 28. Wang JP, Lin M, Xu Z, Zhang YH, Shi ZQ, Qian JM, et al. Microstructure and
TE D
mechanical properties of C/C–SiC composites fabricated by a rapid processing method. J Eur Ceram Soc 2009;29(14):3091-7.
29. Padture NP. Advanced structural ceramics in aerospace propulsion. Nat Mater.
EP
2016;15(8):804-9.
30. Krenkel W, Heidenreich B, Renz R. C/C–SiC Composites for Advanced Friction
AC C
Systems. Adv Eng Mater 2002;4(7):427-36. 31. Stadler Z, Krnel K, Kosmac T. Friction behavior of sintered metallic brake pads on a C/C–SiC composite brake disc. J Eur Ceram Soc 2007;27(2–3):1411-7. 32. Xu YD, Zhang YN, Cheng LF, Zhang LT, Lou JJ, Zhang JZ. Preparation and friction behavior of carbon fiber reinforced silicon carbide matrix composites. Ceram Int 2007;33(3):439-45. 16
ACCEPTED MANUSCRIPT
33. Liu L, Li HJ, Feng W, Shi XH, Wu H, Zhu JL. Effect of surface ablation products on the ablation resistance of C/C–SiC composites under oxyacetylene torch. Corros Sci
RI PT
2013;67:60-6. 34. Feng B, Li HJ, Zhang YL, Liu L, Yan M. Effect of SiC/ZrC ratio on the mechanical and ablation properties of C/C–SiC–ZrC composites. Corros Sci 2014;82:27-35.
SC
35. Xiong CY, Li TH, Zhao TK, Khan M, Wang JQ, Ji XL, et al. Preparation of C/C– SiC composite by low temperature compression molding–liquid silicon infiltration and
M AN U
its application in automobile brake. Ceram Int 2016;42(1, Part B):1057-62. 36. Naslain R. Design, preparation and properties of non-oxide CMCs for application in engines and nuclear reactors: an overview. Compos Sci Technol 2004;64(2):155-70. 37. Jia Y, Li KZ, Xue LZ, Ren JJ, Zhang SY, Li HJ. Mechanical and electromagnetic
TE D
shielding performance of carbon fiber reinforced multilayered (PyC–SiC)n matrix composites. Carbon 2017;111:299-308.
38. Menard KP. Dynamic mechanical analysis: a practical introduction: CRC press;
EP
2008.
39. Hou XH, Li HJ, Wang C, Zhu ZG, Shen J. Internal friction behavior of carbon–
AC C
carbon composites. Carbon 2000;38(15):2095-101. 40. Zhang YH, Xiao ZC, Wang JP, Yang JF, Jin ZH. Effect of pyrocarbon content in C/C preforms on microstructure and mechanical properties of the C/C–SiC composites. Mater Sci Eng A 2009;502(1–2):64-9. 41. Zener C. Elasticity and anelasticity of metals: University of Chicago press; 1948. 42. Granato A, Lücke K. Theory of Mechanical Damping Due to Dislocations. J Appl 17
ACCEPTED MANUSCRIPT
Phys 1956;27(6):583-93. 43. KÊ TS. Anomalous Internal Friction Associated with the Precipitation of Copper in
RI PT
Cold-Worked Al-Cu Alloys. Phys Rev 1950;78(4):420-3. 44. Fukuhara M, Abe Y. High-temperature elastic moduli and internal frictions of α-SiC ceramic. J Mater Sci Lett 1993;12(9):681-3.
SC
45. Jia Y, Li KZ, Xue LZ, Ren JJ, Zhang SY, Zhang XM. Microstructure and
mechanical properties of carbon fiber reinforced multilayered (PyC–SiC)n matrix
AC C
EP
TE D
M AN U
composites. Mater Des 2015;86:55-60.
18
ACCEPTED MANUSCRIPT
Figure Captions Fig. 1 Schematic diagram of preform architecture (a) and deposited layer structure (b) of
RI PT
C/(PyC–SiC)n composites. Fig. 2 BSE images of polished cross-sectional morphology and EDS mapping of CS1 (a), CS2 (b) and CS4 (c), insets are enlarged drawings of dotted boxes.
SC
Fig. 3 PLM micrographs of CS1 (a), CS2 (b) and CS4 (c).
Fig. 4 Internal friction of C/(PyC–SiC)n composites vs. frequency at strain amplitude of
M AN U
9 µm and RT.
Fig. 5 Internal friction of C/(PyC–SiC)n composites vs. strain amplitude at frequency of 5 Hz and RT.
Fig. 6 Internal friction of C/(PyC–SiC)n composites vs. temperature at frequency of 5
TE D
Hz and strain amplitude of 9 µm.
Fig. 7 Internal friction of CS2 versus frequency before and after damage at strain amplitude of 9 µm and RT.
EP
Fig. 8 Schematic of stress distribution in C/(PyC–SiC)n composites during internal friction tests.
AC C
Fig. 9 SEM image of defects in CS2 after slight damage.
19
AC C
EP
TE D
M AN U
SC
Fig. 1
RI PT
ACCEPTED MANUSCRIPT
Fig. 2 20
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
EP
TE D
Fig. 3
Fig. 5
AC C
Fig. 4
Fig. 6
Fig. 7 21
M AN U
SC
Fig. 8
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Fig. 9
22
S1359-8368(16)32576-8
DOI:
10.1016/j.compositesb.2017.01.068
Reference:
JCOMB 4879
To appear in:
Composites Part B
Received Date: 4 November 2016 Revised Date:
9 December 2016
Accepted Date: 28 January 2017
Please cite this article as: Jia Y, Li K, Xue L, Ren J, Zhang S, Internal friction behaviour of carbon fibre reinforced multilayered (PyC–SiC)n matrix composites, Composites Part B (2017), doi: 10.1016/ j.compositesb.2017.01.068. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Internal friction behaviour of carbon fibre reinforced multilayered (PyC–SiC)n matrix composites
RI PT
Yan Jia, Kezhi Li∗, Lizhen Xue, Junjie Ren, Shouyang Zhang† State Key Laboratory of Solidification Processing, Carbon/carbon composites Research Centre, Northwestern Polytechnical University, Xi’an 710072, PR China
SC
Abstract
Carbon fibre reinforced multilayered (PyC–SiC)n matrix (C/(PyC–SiC)n)
M AN U
composites were prepared by alternate deposition of pyrocarbon (PyC) and SiC inside preforms via chemical vapour infiltration. The matrix microstructures and internal friction behaviours of C/(PyC–SiC)n composites (n=1, 2 and 4) under different testing conditions of frequency, strain amplitude and temperature were studied. The results
TE D
show that with increasing the number of sequences (n), the internal friction increases due to the enhancement of interfacial internal friction. The internal friction of C/(PyC– SiC)n composites increases with the increase of frequency related to thermoelastic
EP
mechanism, but exhibits anomalous amplitude effect in the testing amplitude range. Effect of temperature on internal friction behaviours is attributed to combined effects of
AC C
carbon fibres, PyC–SiC matrices and interfaces between fibres and PyC, adjoining PyC and SiC layers. It is also found that internal friction is sensitive to microstructural defects induced by damage. These results indicate that internal friction is an effective and efficient method to characterize the structural evolution and internal damage of C/(PyC–SiC)n composites non-destructively. * Corresponding author. Tel.: +86 29 88495764; E-mail address: [email protected] (K. Li) † Corresponding author. Tel.: +86 29 88492272; E-mail address: [email protected] (S. Zhang) 1
ACCEPTED MANUSCRIPT
Keywords: A. Carbon–carbon composites (CCCs); A. Layered structures; B. Internal friction/damping; E. Chemical vapour deposition (CVD)
RI PT
1. Introduction Internal friction or damping capacity of a material refers to its ability to convert mechanical vibration energy to internal energy irreversibly, which is an important
SC
parameter in mechanical design [1]. Internal friction depends on the constitutions and
microstructure of materials and is very sensitive to the variation of microstructure [2].
M AN U
Importantly, it is a non-destructive detection method as an effective and efficient tool to study microstructural evolution within materials [3]. Thus, internal friction not only plays an enabling role in evaluating dynamic mechanical properties, but also is a basic approach for characterizing structural changes. Up to now, enormous work has been
TE D
carried out to internal friction analysis for studying composites, including metal-based [4-8], polymer-based [9-12], carbon-based [13-18] and SiC-based composites [19-23]. Various mechanisms and models have been established to predict the internal friction of
EP
composite materials [24-27].
Carbon fibre reinforced carbon–silicon carbide matrix (C/C–SiC) composites have
AC C
been widely used as one of the most promising structural materials due to their superior properties, such as low density, high strength at elevated temperature and excellent friction performance [28-30]. C/C–SiC composites are also potential candidates for brake discs and clutch systems [31], which involve structural vibration and internal friction. Therefore, it is necessary to study the behaviour of internal friction of C/C–SiC composites. 2
ACCEPTED MANUSCRIPT
Currently, many preparation processes have been developed to manufacture C/C– SiC composites, including chemical vapour infiltration (CVI) [32], polymer
RI PT
impregnation pyrolysis (PIP) [33, 34] and liquid silicon infiltration (LSI) [35]. It is well known that PIP process would inevitably produce many defects resulting from gas
volatilization and shrinkage within matrix during pyrolysis process and LSI process
SC
requires high temperature (above 1410 °C) for reaction of C and Si leading to great
residual stress and matrix cracking [36], which can severely deteriorate the performance
M AN U
of C/C–SiC composites. Inversely, C/C–SiC composites prepared by CVI have a good matrix and controlled microstructure [36], indicating that CVI is an effective way to obtain C/C–SiC composites with high performance.
In our previous work [37], a new member of C/C–SiC composites was fabricated
TE D
by alternate deposition of pyrocarbon (PyC) and SiC inside preforms via isothermal CVI (ICVI), i.e. carbon fibre reinforced multilayered (PyC–SiC)n matrix (C/(PyC–SiC)n) composites. Static mechanical and electromagnetic interference shielding properties of
EP
the composites were studied intensively. However, the fundamental internal friction of C/(PyC–SiC)n composites has yet to be reported. In this paper, the internal friction
AC C
behaviours of C/(PyC–SiC)n composites with various microstructures were investigated.
The internal friction mechanisms under different testing conditions were also discussed in detail.
2. Experimental procedure 2.1 Materials preparation 2.5D needle-punched carbon fibre felts (Yixing Tianniao High Technology Co. Ltd., 3
ACCEPTED MANUSCRIPT
China) were used as preforms. The fibre type was commercial 12 k T700 PAN-based (Toray, Japan) carbon fibre with a filament diameter of 7 µm, and fibre volume fraction
RI PT
of the preform was about 25%. Schematic diagram of preform architecture is shown in Fig. 1(a). It was fabricated by repeatedly overlapping layers of 0° non-woven carbon fibre cloth, short-cut fibre web, 90° non-woven carbon fibre cloth and short-cut fibre
SC
web with needle-punching step by step. ICVI was employed to alternately infiltrate PyC and SiC into preforms. One PyC–SiC sequence ((PyC–SiC)1), two PyC–SiC sequences
M AN U
((PyC–SiC)2) and four PyC–SiC sequences ((PyC–SiC)4) were deposited on carbon fibres to prepare three kinds of C/(PyC–SiC)n composites (designated as CS1, CS2 and CS4, respectively), as illustrated in Fig. 1(b). PyC matrix was deposited at 1100 °C under a negative pressure of 5 kPa with the flow rate of methane at 1.2 L/min. As for
TE D
the deposition process of SiC matrix, methyltrichlorosilane (MTS, CH3SiCl3) was used as precursor and carried by bubbling hydrogen in gas phase, and argon was employed as dilute gas. The deposition temperature was controlled at 1100 °C under a pressure of 3
EP
kPa, molar ratio of H2 to MTS was 10, and flow rate of Ar was 300 mL/min. Further details of deposition time for each layer of PyC and SiC were described elsewhere [37].
AC C
The final bulk density and open porosity of each composite were 1.34 g/cm3 and 35.1%,
1.33 g/cm3 and 37.2%, 1.36 g/cm3 and 34.6%, respectively. 2.2 Characterization
The internal friction tests were performed on a dynamic mechanical analyser (DMA, Q800, TA, USA) by mode of three-point bending force vibration. Samples were cut into rectangular bars with dimensions of 40 mm × 5 mm × 3 mm. The loading 4
ACCEPTED MANUSCRIPT
direction was perpendicular to the non-woven cloth. The testing conditions were given as follows: frequency was from 0.1 to 10 Hz at strain amplitude of 9 µm and room
RI PT
temperature (RT), strain amplitude varied from 5 to 10 µm at frequency of 5 Hz and RT, temperature range was from RT to 300 °C with a heating rate of 5 °C/min at frequency of 5 Hz and strain amplitude of 9 µm. All runs were conducted in air. The internal
SC
friction (Q-1) was calculated according to the following equation [38]: Q-1 = tan δ = E''/E'
(1)
M AN U
where δ is the loss angle between applied stress and strain, E' is the dynamic storage (real) modulus and E'' is the dynamic loss (imaginary) modulus.
Microstructures of C/(PyC–SiC)n composites were observed under polarized light microscope (PLM, DMLP, Leica, Germany). Polished cross-sectional morphology of
TE D
the composites was characterized by scanning electron microscope (SEM, VEGA3, Tescan, Czech) equipped with energy dispersive spectroscope (EDS, Oxford INCA, UK).
EP
3. Results
3.1 Microstructure characterization
AC C
Fig. 2 shows the backscattered electron (BSE) images of polished cross-sectional
morphology and EDS mapping of insets of C/(PyC–SiC)n composites (n=1, 2 and 4). It
can be seen that the matrices of C/(PyC–SiC)n composites are composed of binary phases displaying dark grey and light grey. Additionally, the matrices become increasingly fine and close in texture with the increase of PyC–SiC sequences number (n). Under short deposition time of 40 h, there remain a large number of inter-bundle 5
ACCEPTED MANUSCRIPT
and intra-bundle pores. Enlarged views of dotted boxes (insets of Fig. 2(a)-(c)) display that the dark grey phase is carbon and light grey phase is SiC, attesting by
RI PT
corresponding EDS mapping of insets. It can be clearly observed from the insets that CS1, CS2 and CS4 have a two-layer, four-layer and eight-layer structure around the carbon fibres. Owing to the effective ICVI process, PyC and SiC are sequentially
SC
well-infiltrated into the preforms and formed multilayered matrices as designed.
Fig. 3 presents the PLM micrographs of CS1, CS2 and CS4, which show various
M AN U
types of matrix microstructures. The multilayered matrices exhibit alternately black and white concentric annuluses marked by arrows, which could be attributed to different light reflectivity under PLM. PyC has high optical activity displaying white, whereas SiC is insensitive to the polarized light appearing black. It is worth noting that PLM
TE D
images of layered microstructures consist with BSE images (insets of Fig. 2) apart from different indicative colours of PyC and SiC. 3.2 Internal friction
EP
3.2.1 Internal friction versus frequency
Internal friction of C/(PyC–SiC)n composites as a function of frequency at strain
AC C
amplitude of 9 µm and RT is illustrated in Fig. 4. Among the three composites, CS4 has
the highest internal friction and CS1 has the lowest internal friction in the whole testing frequency. The internal friction of all composites is almost unchanged from 0.1 to 2 Hz, and then begins to increase until 10 Hz. The similar relationship between internal friction and frequency is also reported by Hou et al. [14] and Cheng et al. [17] in C/C composites. The results indicate that high frequency could affect the internal friction 6
ACCEPTED MANUSCRIPT
behaviours of C/(PyC–SiC)n composites. 3.2.2 Internal friction versus strain amplitude
RI PT
Fig. 5 displays internal friction of C/(PyC–SiC)n composites versus strain amplitude at frequency of 5 Hz and RT. It is obvious that the internal friction is
dependent on testing amplitude. As the strain amplitude increases from 5 to 10 µm, the
SC
internal friction of all composites decreases gradually. It is also observed that the overall level of internal friction in CS4 is higher than that in CS1 and CS2.
M AN U
3.2.3 Internal friction versus temperature
Fig. 6 shows the dependence of internal friction on temperature at frequency of 5 Hz and strain amplitude of 9 µm for C/(PyC–SiC)n composites. The internal friction values are in the sequences of CS4 > CS2 > CS1 over the whole testing temperature
TE D
range. Moreover, all internal friction-temperature curves can be divided into three stages with the increase of temperature. The internal friction declines monotonically from RT to about 50 °C in stage I. In stage II, the internal friction curves of CS1 and
EP
CS2 keeps almost constant from 50 to 200 °C in spite of gentle fluctuations, while the
AC C
curve of CS4 fluctuates greatly. When temperature is above 200 °C in stage III, the internal friction of all composites decreases again with varying slopes. Especially, two internal friction peaks appear at about 75 and 240 °C for CS4 in stage II and III, respectively. The similar phenomena are also found in C/C [39], SiC/SiC [19] and C/SiC [21] composites except for the different peak temperatures. 3.2.4 Internal friction versus frequency before and after damage Fig. 7 presents the internal friction of CS2 versus frequency before and after slight 7
ACCEPTED MANUSCRIPT
damage, which is conducted by a mild three-point bending. The obtained samples still have highly structural integrity and cannot be distinguished from other normal samples
RI PT
through appearance. As shown in Fig. 7, the internal friction after slight damage is much higher than that before damage, which demonstrates that the internal friction is very sensitive to damage.
SC
4. Discussion 4.1 Intrinsic internal friction
M AN U
It has been recognized that the internal friction behaviour of a composite is determined not only by matrix and fibres, but also by properties of interfaces within the composite [40]. Internal friction (Q-1) is the combined results of such various factors, which can be described as follows:
(2)
TE D
Q-1 = tan δf + tan δm + tan δint
where tan δf, tan δm and tan δint are the internal friction caused by fibres, matrix and interfaces, respectively.
EP
For all three composites, they have the same carbon felt preforms, similar density, open porosity, PyC and SiC content [37]. In addition, the interfaces between carbon
AC C
fibres and PyC are determined by the volume fraction of fibres in C/(PyC–SiC)n
composites [15]. Thus, the contributions of carbon fibres, PyC-SiC matrix and fibre/PyC interfaces to internal friction are approximately equal among them. The difference of internal friction can be attributed to interfaces between PyC and SiC (PyC/SiC) within the matrices. Eq. (2) can be represented as follows: Q-1 = tan δf + tan δm + tan δint, 1 + tan δint, 2 = tan δs + tan δint, 2 8
(3)
ACCEPTED MANUSCRIPT
where tan δint, 1 and tan δint, 2 are internal friction resulting from fibre/PyC and PyC/SiC interfaces, respectively. tan δs is the sum of tan δf, tan δm and tan δint, 1.
RI PT
In general, weak or strong interfacial bonding both can contribute to the internal friction. Many researchers have found that interfacial internal friction (tan δint) increases with increasing fibres or particulates volume fraction [4, 14, 15]. Obviously, as volume
SC
fraction of reinforcement increases, the number of reinforcement/matrix interfaces
increases. As shown in Fig. 2 and 3, the number of PyC/SiC interfaces increases with
M AN U
the increase of sequences number (n), leading to tan δint, 2 (CS4) > tan δint, 2 (CS2) > tan δint, 2 (CS1). Combined with previous analysis, it can be concluded that Q-1 (CS4) > Q-1 (CS2) > Q-1 (CS1) regardless of external influences. This could explain the growing trend of internal friction in Fig. 4-6 nicely.
TE D
4.2 Effects of frequency and amplitude on internal friction
In the experiments, the testing samples are forcedly vibrated under flexural stress, which could generate cyclic heat flow from the region of compressive stress to the
EP
region of tensile stress in the inhomogeneous composites (Fig. 8) [25, 26]. The temperature gradients are produced resulting in dissipation of mechanical energy.
AC C
Therefore, the thermoelastic mechanism should be given priority as the most fundamental internal friction mechanisms [1, 25]. According to the approximate relationship between frequency and internal friction proposed by Zener [41], thermoelastic internal friction would increase with increasing frequency and could be ignored at low frequency, which is in a good agreement with the result in Fig. 4. Moreover, the thermoelastic internal friction is frequency-dependent and 9
ACCEPTED MANUSCRIPT
amplitude-independent [39]. However, as shown in Fig. 4 and 5, the internal friction of C/(PyC–SiC)n composites is influenced by not only frequency but also amplitude. It is
RI PT
reasonable to suppose that there exist other mechanisms affecting the internal friction behaviours of C/(PyC–SiC)n composites.
Theoretically, the amplitude dependence is usually relevant to static hysteresis [38].
SC
As opposed to thermoelastic mechanism, it is independent of frequency and dependent on amplitude [1]. During measurement, the stress-strain curves of C/(PyC–SiC)n
M AN U
composites will give hysteresis loop under cyclic stress, which can be explained by the dislocation mechanism (K-G-L theory) proposed by Granato and Lücke [42]. The area enclosed by loop corresponds to the dissipated energy. In such circumstance, the internal friction would increase with the increase of strain amplitude. But in the present
TE D
work, the internal friction of all three composites reduces gradually as amplitude increases (Fig. 5), presenting anomalous amplitude effect [43]. The results demonstrate that other mechanisms are coupled with static hysteresis mechanism in terms of
EP
amplitude-dependence, which remains to be furtherly studied. 4.3 Effect of temperature on internal friction
AC C
The variation of internal friction-temperature curves becomes complicated
compared with internal friction-frequency and internal friction-strain amplitude curves, which could be attributed to the specific characteristics of components (fibre, matrix and interface) versus temperature. The internal friction of C/C composites and graphite has been found to decrease with increasing temperature [16, 39]. On the contrary, the internal friction of CVI SiC/SiC composites and SiC increase as temperature increases 10
ACCEPTED MANUSCRIPT
[19, 44]. It can be concluded that the different variation trends are intrinsic properties of carbon and SiC materials. Furthermore, the thermal stress and interfacial delamination
RI PT
are induced due to the mismatches of coefficient of thermal expansion (CTE) between carbon fibres and PyC, PyC and SiC during the cooling stage [45]. Hence, the overall
internal friction behaviours can be understood by taking into account all factors, which
SC
may compete with each other.
In low temperature range (stage I), the carbon fibres and PyC–SiC matrix are
M AN U
predominant and the internal friction is controlled by the behaviours of carbon and SiC. The carbon content is higher than SiC content [37] and the internal friction of carbon (5~8 × 10-3 [22]) is also larger than that of SiC (1.2~5 × 10-3 [22]). Thus, carbon material plays a dominant role, resulting in the decline of internal friction. When
TE D
temperature rises, the thermal stress at fibre/matrix and PyC/SiC interfaces can be released and delamination may be reduced or sealed gradually, which will improve the bonding of fibre/matrix interfaces and adjoining PyC and SiC layers. The attrition
EP
energy loss increases leading to the enhancement of internal friction. On the other hand, the character of SiC material also contributes to the increase of internal friction in high
AC C
temperature. The combined positive contributions could compensate the negative contribution of carbon material character, so the internal friction of CS1 and CS2 remains nearly stable in stage II. As temperature continues to rise (stage III), the internal friction produced by different kinds of defects including dislocation, debonding and microcracks would decrease due to the decrement of internal friction sources according to the dislocation mechanism [42]. Therefore, the overall internal friction decreases. 11
ACCEPTED MANUSCRIPT
Compared with CS1 and CS2, the occurrence of two peaks in CS4 could be the competitive results between a greater deal of interfacial internal friction and character of
RI PT
carbon itself. 4.4 Effect of damage on internal friction
Fig. 9 shows the SEM image of defects in CS2 after slight damage. Though CS2
SC
endures slight damage, matrix microcracks and debonding between fibres and PyC are
generated, which is relevant to the significant differences of internal friction before and
M AN U
after damage (Fig. 7). Under cyclic stress, such defects could dissipate much energy resulting in sharp increase of internal friction. Therefore, internal friction is a good indicator to evaluate the occurrence of damage, especially micro-damage. 5. Conclusions
TE D
In the present work, internal friction behaviours of C/(PyC–SiC)n composites with different number of PyC–SiC sequences (n=1, 2 and 4) were studied under testing conditions of frequency, strain amplitude and temperature, respectively. The sequences
EP
number (n) exerts an important influence on the internal friction of C/(PyC–SiC)n composites. With increasing the sequences number (n), the interfacial internal friction
AC C
increases resulting in Q-1 (CS4) > Q-1 (CS2) > Q-1 (CS1) under different testing conditions. The internal friction keeps unchanged at lower frequency and increases at higher frequency range, which can be explained by thermoelastic mechanism. The internal friction of all composites decreases in the whole testing strain amplitude, but the mechanisms of anomalous amplitude effect are still not clear. The internal friction-temperature dependence is controlled by interfacial characteristic and intrinsic 12
ACCEPTED MANUSCRIPT
properties of carbon and SiC, exhibiting different trends in three stages. Moreover, the superior responsiveness of internal friction to damage demonstrates that it is a
RI PT
non-destructive detection method to determine the internal damage of materials effectively and efficiently. Acknowledgements
SC
This work was supported by the National Natural Science Foundation of China (Grant Nos. 51521061 and 51472202) and “111” Project (Grant No.B08040).
1.
M AN U
References
Blanter MS, Golovin I, Neuhauser H, Sinning H. Internal friction in metallic
materials. A handbook Series: Springer Series in Materials Science, 2007. 2.
Berthelot J-M, Assarar M, Sefrani Y, Mahi AE. Damping analysis of composite
3.
TE D
materials and structures. Compos Struct 2008;85(3):189-204.
Boccaccini AR, Ponton CB, Chawla KK. Development and healing of matrix
microcracks in fibre reinforced glass matrix composites: assessment by internal friction.
4.
EP
Mater Sci Eng A 1998;241(1–2):141-50.
Perez RJ, Zhang J, Gungor MN, Lavernia EJ. Damping behavior of 6061Al/Gr
AC C
metal matrix composites. Metall Trans A 1993;24(3):701-12. 5.
Parrini L, Schaller R. Thermal stresses in metal matrix composites studied by
internal friction. Acta Mater 1996;44(12):4881-8. 6.
Zhang HX, Gu MY. Internal friction behavior in SiC particulate reinforced
aluminum metal matrix composite in thermal cycling. J Alloy Compd 2006;426(1– 2):247-52. 13
ACCEPTED MANUSCRIPT
7.
Hu J, Wu GH, Zhang Q, Gou HS. Mechanical properties and damping capacity of
SiCp/TiNif/Al composite with different volume fraction of SiC particle. Compos Part B:
8.
RI PT
Eng 2014;66:400-6. Shin J, Choi K, Shiko S, Choi H, Bae D. Mechanical damping behavior of
Al/C60-fullerene composites with supersaturated Al–C phases. Compos Part B: Eng
9.
SC
2015;77:194-8.
Feng JM, Guo ZS. Effects of temperature and frequency on dynamic mechanical
M AN U
properties of glass/epoxy composites. J Mater Sci 2016;51(5):2747-58.
10. Babal AS, Singh BP, Jyoti J, Sharma S, Arya AK, Dhakate SR. Synergistic effect on static and dynamic mechanical properties of carbon fiber-multiwalled carbon nanotube hybrid polycarbonate composites. RSC Adv 2016;6(72):67954-67.
TE D
11. Rafiee F, Otadi M, Goodarzi V, Khonakdar HA, Jafari SH, Mardani E, et al. Thermal and dynamic mechanical properties of PP/EVA nanocomposites containing organo-modified layered double hydroxides. Compos Part B: Eng 2016;103:122-30.
EP
12. Ashworth S, Rongong J, Wilson P, Meredith J. Mechanical and damping properties of resin transfer moulded jute-carbon hybrid composites. Compos Part B: Eng
AC C
2016;105:60-6.
13. Qiao SR, Zhou SR, Bai SH, Sun XY. The internal friction of unidirectional CC composites fabricated in a magnetic field. Carbon 1997;35(3):389-92. 14. Hou XH, Li HJ, Shen J, Wang C, Zhu ZG. Effects of microstructure on the internal friction of carbon–carbon composites. Mater Sci Eng A 2000;286(2):250-6. 15. Wang C, Zhu ZG, Hou XH, Li HJ. Damping characteristics of CVI-densified 14
ACCEPTED MANUSCRIPT
carbon–carbon composites. Carbon 2000;38(13):1821-4. 16. Kogo Y, Iijima Y, Igata N, Ota Ki. Internal friction of carbon–carbon composites at
RI PT
elevated temperatures. J Alloy Compd 2003;355(1–2):148-53. 17. Cheng J, Li HJ, Zhang SY, Xue LZ, Luo WF. Internal friction behavior of
unidirectional carbon/carbon composites after different fatigue cycles. Mater Sci Eng A
SC
2014;600:129-34.
18. Liu Y, Fu QG, Zhao FL, Sun GD, Li HJ. Internal friction vs. thermal shock in C/C
M AN U
composites. Compos Part B: Eng 2016;106:59-65.
19. Sato S, Serizawa H, Araki H, Noda T, Kohyama A. Temperature dependence of internal friction and elastic modulus of SiC/SiC composites. J Alloy Compd 2003;355(1–2):142-7.
TE D
20. Zhang J, Xu YD, Zhang LT, Cheng LF. Internal friction and dynamic modulus of three-dimensional silicon carbide-matrix composites. Mater Lett 2005;59(19– 20):2535-8.
EP
21. Zhang Q, Cheng LF, Wang W, Zhang LT, Xu YD. Effect of SiC coating and heat treatment on damping behavior of C/SiC composites. Mater Sci Eng A 2008;473(1–
AC C
2):254-8.
22. Hou JT, Qiao SR, Lu GF, Zhang CY, Zhang YB. Influence of heat treatment on the internal
friction
of
2D-C/SiC
composites.
J
Mater
Process
Technol.
2009;209(7):3555-60. 23. Hong ZL, Cheng LF, Zhang LT, Wang YG. Internal Friction Behavior of C/SiC Composites with Environmental Barrier Coatings in Corrosive Environment. Int J Appl 15
ACCEPTED MANUSCRIPT
Ceram Tec 2011;8(2):342-50. 24. Chandra R, Singh SP, Gupta K. Micromechanical damping models for
RI PT
fiber-reinforced composites: a comparative study. Compos Part A: Appl Sci Manuf 2002;33(6):787-96.
25. Chandra R, Singh SP, Gupta K. Damping studies in fiber-reinforced composites-a
SC
review. Compos Struct 1999;46(1):41-51.
26. Chandra R, Singh SP, Gupta K. A study of damping in fiber-reinforced composites.
M AN U
J Sound Vib 2003;262(3):475-96.
27. Treviso A, Van Genechten B, Mundo D, Tournour M. Damping in composite materials: Properties and models. Compos Part B: Eng 2015;78:144-52. 28. Wang JP, Lin M, Xu Z, Zhang YH, Shi ZQ, Qian JM, et al. Microstructure and
TE D
mechanical properties of C/C–SiC composites fabricated by a rapid processing method. J Eur Ceram Soc 2009;29(14):3091-7.
29. Padture NP. Advanced structural ceramics in aerospace propulsion. Nat Mater.
EP
2016;15(8):804-9.
30. Krenkel W, Heidenreich B, Renz R. C/C–SiC Composites for Advanced Friction
AC C
Systems. Adv Eng Mater 2002;4(7):427-36. 31. Stadler Z, Krnel K, Kosmac T. Friction behavior of sintered metallic brake pads on a C/C–SiC composite brake disc. J Eur Ceram Soc 2007;27(2–3):1411-7. 32. Xu YD, Zhang YN, Cheng LF, Zhang LT, Lou JJ, Zhang JZ. Preparation and friction behavior of carbon fiber reinforced silicon carbide matrix composites. Ceram Int 2007;33(3):439-45. 16
ACCEPTED MANUSCRIPT
33. Liu L, Li HJ, Feng W, Shi XH, Wu H, Zhu JL. Effect of surface ablation products on the ablation resistance of C/C–SiC composites under oxyacetylene torch. Corros Sci
RI PT
2013;67:60-6. 34. Feng B, Li HJ, Zhang YL, Liu L, Yan M. Effect of SiC/ZrC ratio on the mechanical and ablation properties of C/C–SiC–ZrC composites. Corros Sci 2014;82:27-35.
SC
35. Xiong CY, Li TH, Zhao TK, Khan M, Wang JQ, Ji XL, et al. Preparation of C/C– SiC composite by low temperature compression molding–liquid silicon infiltration and
M AN U
its application in automobile brake. Ceram Int 2016;42(1, Part B):1057-62. 36. Naslain R. Design, preparation and properties of non-oxide CMCs for application in engines and nuclear reactors: an overview. Compos Sci Technol 2004;64(2):155-70. 37. Jia Y, Li KZ, Xue LZ, Ren JJ, Zhang SY, Li HJ. Mechanical and electromagnetic
TE D
shielding performance of carbon fiber reinforced multilayered (PyC–SiC)n matrix composites. Carbon 2017;111:299-308.
38. Menard KP. Dynamic mechanical analysis: a practical introduction: CRC press;
EP
2008.
39. Hou XH, Li HJ, Wang C, Zhu ZG, Shen J. Internal friction behavior of carbon–
AC C
carbon composites. Carbon 2000;38(15):2095-101. 40. Zhang YH, Xiao ZC, Wang JP, Yang JF, Jin ZH. Effect of pyrocarbon content in C/C preforms on microstructure and mechanical properties of the C/C–SiC composites. Mater Sci Eng A 2009;502(1–2):64-9. 41. Zener C. Elasticity and anelasticity of metals: University of Chicago press; 1948. 42. Granato A, Lücke K. Theory of Mechanical Damping Due to Dislocations. J Appl 17
ACCEPTED MANUSCRIPT
Phys 1956;27(6):583-93. 43. KÊ TS. Anomalous Internal Friction Associated with the Precipitation of Copper in
RI PT
Cold-Worked Al-Cu Alloys. Phys Rev 1950;78(4):420-3. 44. Fukuhara M, Abe Y. High-temperature elastic moduli and internal frictions of α-SiC ceramic. J Mater Sci Lett 1993;12(9):681-3.
SC
45. Jia Y, Li KZ, Xue LZ, Ren JJ, Zhang SY, Zhang XM. Microstructure and
mechanical properties of carbon fiber reinforced multilayered (PyC–SiC)n matrix
AC C
EP
TE D
M AN U
composites. Mater Des 2015;86:55-60.
18
ACCEPTED MANUSCRIPT
Figure Captions Fig. 1 Schematic diagram of preform architecture (a) and deposited layer structure (b) of
RI PT
C/(PyC–SiC)n composites. Fig. 2 BSE images of polished cross-sectional morphology and EDS mapping of CS1 (a), CS2 (b) and CS4 (c), insets are enlarged drawings of dotted boxes.
SC
Fig. 3 PLM micrographs of CS1 (a), CS2 (b) and CS4 (c).
Fig. 4 Internal friction of C/(PyC–SiC)n composites vs. frequency at strain amplitude of
M AN U
9 µm and RT.
Fig. 5 Internal friction of C/(PyC–SiC)n composites vs. strain amplitude at frequency of 5 Hz and RT.
Fig. 6 Internal friction of C/(PyC–SiC)n composites vs. temperature at frequency of 5
TE D
Hz and strain amplitude of 9 µm.
Fig. 7 Internal friction of CS2 versus frequency before and after damage at strain amplitude of 9 µm and RT.
EP
Fig. 8 Schematic of stress distribution in C/(PyC–SiC)n composites during internal friction tests.
AC C
Fig. 9 SEM image of defects in CS2 after slight damage.
19
AC C
EP
TE D
M AN U
SC
Fig. 1
RI PT
ACCEPTED MANUSCRIPT
Fig. 2 20
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
EP
TE D
Fig. 3
Fig. 5
AC C
Fig. 4
Fig. 6
Fig. 7 21
M AN U
SC
Fig. 8
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Fig. 9
22