- Email: [email protected]
The effect of microstructural evolution on micromechanical behavior of pyrolytic carbon after heat treatment
Journal Pre-proof The effect of microstructural evolution on micromechanical behavior of pyrolytic carbon after heat treatment
Tiyuan Wang, Hejun Li, Shouyang Zhang, Kun Li, Wei Li PII:
S0925-9635(19)30818-0
DOI:
https://doi.org/10.1016/j.diamond.2020.107729
Reference:
DIAMAT 107729
To appear in:
Diamond & Related Materials
Received date:
22 November 2019
Revised date:
9 January 2020
Accepted date:
22 January 2020
Please cite this article as: T. Wang, H. Li, S. Zhang, et al., The effect of microstructural evolution on micromechanical behavior of pyrolytic carbon after heat treatment, Diamond & Related Materials (2020), https://doi.org/10.1016/j.diamond.2020.107729
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. 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.
© 2020 Published by Elsevier.
Journal Pre-proof
Title page
Title: The effect of microstructural evolution on micromechanical behavior of pyrolytic carbon after heat treatment
Author names and affiliations: Tiyuan Wang, Hejun Li, Shouyang Zhang, Kun Li,
of
Wei Li (State Key Laboratory of Solidification Processing, Northwestern Polytechnical
ro
University, Xi’an, Shaanxi 710072, PR China)
-p
Corresponding author: Hejun Li (Tel.: +86-29-88495004. E-mail address:
lP
Jo ur
na
Declarations of interest: none
re
[email protected])
Journal Pre-proof
The effect of microstructural evolution on micromechanical behavior of pyrolytic carbon after heat treatment Tiyuan Wang, Hejun Li*, Shouyang Zhang*, Kun Li, Wei Li State Key Laboratory of Solidification Processing, Northwestern Polytechnical University, Xi’an, Shaanxi 710072, PR China Abstract
of
In order to investigate the effect of microstructural evolution on the micromechanical behavior of
ro
pyrolytic carbon (PyC), the microstructural parameters and evolution of high texture (HT) and medium texture (MT) PyC after heat treatment from 1600 °C to 2450 °C were studied comparatively. The
-p
micro-indentation test was also performed to reveal the micromechanics. In terms of microstructure, after
re
heat treatment, the HT matrix was directly delaminated into many nano-layers accompanied by multiple microcracks, while the MT matrix was delaminated into micro-layers including lots of cracks. Meanwhile,
lP
numerous nanopores were still observed in the MT matrix after heat treatment at 2450 °C. The interfaces of two textures became smoother and wider, and the width of HT interface was narrower than that of MT
na
at different heat treatment temperatures. Moreover, the increasing tendency of the basal plane length (La) and the height of layered stacking (Lc) of HT were greater than those of MT as HT PyC contained more
Jo ur
C=C bonds than MT PyC. The micro-indentation results showed that the indentation depth of HT was always deeper than that of MT, and the corresponding micro-hardness and elastic modulus of MT were also greater than that of HT after different temperature treatment. Key words: Pyrolytic carbon; Microstructure; Micro-hardness; Lattice structure; Interface 1 Introduction Carbon/Carbon (C/C) composites exhibit superior high temperature performance including low density, high specific strength, high fracture toughness, and outstanding thermal shock resistance, which makes them widely used as throat linings, aircraft brake discs, and nose cones [1-5]. Chemical vapor
*
Corresponding author. Tel.: +86 029 88495004; E-mail address: [email protected] (H. Li)
*
Corresponding author. E-mail addresses: [email protected] (S. Zhang).
Journal Pre-proof
infiltration (CVI) is the most common process for fabricating C/C composites. In the view of microstructures, the first classification of different microstructures of pyrolytic carbon was presented by Gray and Cathcart [6]. Bourrat et al. [7] further classified the microstructure of pyrolytic carbon using extinction angle (Ae) of polarized light microscope (PLM) and selected area electron diffraction (SAED) of TEM observation. Up to now, the pyrolytic carbon usually has four typical textures: high texture (HT), medium texture (MT), low texture (LT) and isotropic carbon (ISO), which was proposed by Reznik and Hüttinger [8]. The mechanical properties of C/C composites depend on the microstructures of PyC, the
of
type of carbon fiber and the state of interface bonding between fiber and matrix [9-14]. On the other hand, graphitization, as a post-treatment after the infiltration process, is an essential factor for the mechanics of
ro
C/C composites [15, 16]. So far, numerous studies concentrate on the effect of heat treatment on
-p
mechanical properties, but the theoretical contribution on microstructures in this field is far from enough.
re
The heat treatment temperature has an essential influence on the microstructure and mechanical properties of C/C material. In the early 1970s, Pierson et al. [17] found that the flexural strength and
lP
modulus of C/C composites decreased after high heat treatment temperature (HTT). Piat et al. [18] reported the effect of heat treatment temperature on the mechanical properties of tangential fiber
na
reinforced MT and HT. The results showed that with the increase of heat treatment temperature (2200-2900 °C), the strength of HT material decreased and exhibited typical pseudoplasticity, while the
Jo ur
strength of MT gradually increased and presented obvious brittleness. Ozcan et al. [19] investigated the effect of HTT on interfacial shear strength (IFSS) of C/C composites and detected that compared to heat treatment at 1800 °C, the IFSS of C/C composites decreased after heat treatment at 2100 °C and 2400 °C. This result was in good agreement with another investigation [20], and it can be partially attributed to the reorganization of the graphene sheets in the matrix in the vicinity of fiber/matrix interface. Xia et al. [21] analyzed the microstructure and flexural properties of C/C composites after heat treatment from 1800 °C to 2500 °C. It was found that heat treatment resulted in distinct interfacial cracks and concentric cracks in smooth laminar pyrolytic carbon and the flexural strength decreased rapidly after treated at 2100 °C while increased back at 2450 °C. The reason was that the poor stress transfer capabilities and high stress concentration caused by multiple interfacial debonding and concentric cracks. Li et al. [22] reported that the crystalline thickness, graphitization degree, flexural strength and modulus of C/C composites with HT matrix increased obviously after heat treatment at 2500 °C.
Journal Pre-proof
For a long time, there is a lack of detailed and sufficient analyses about the effect of microstructure evolution on micromechanics of pyrolytic carbon after heat treatment, even though the mechanical properties of C/C composites were probed by many above researches. Thus, a further characterization of the structure of C/C composites in both micro and nano scale is necessary owing to the insufficient correlation between the microstructure and mechanical properties. Moreover, a thorough understanding of their microstructural evolution and lattice parameters after heat treatment is of great significance for the design, application and performance evaluation of the C/C materials. Hence, in this work, the
of
micro-hardness was tested to evaluate the micromechanics of HT and MT under different treatment temperatures, comprehensively considering the delamination, interface, lattice structure morphologies.
ro
Meanwhile, microstructure evolution and bond transformation were also used to explain the difference in
-p
micro-hardness.
re
2 Material and Experiment
lP
2.1 Specimens preparation
2.5D needled PAN-based carbon fiber felts (Yixing Tianniao Co., Ltd., China) with a density of 0.45
na
g/cm3 were used as preforms. Thermal gradient chemical vapor infiltration was employed to infiltrate the preforms at the temperature range of 900 °C ~ 1200 °C, using natural gas (>95%) and argon as the
Jo ur
precursor and carrier gas. The HT and MT C/C composites were finally obtained by controlling the process parameters. The bulk density and porosity of HT and MT C/C composites measured by the Archimedes method were1.78 g/cm3, 8.3%, and 1.75 g/cm3, 8.6% respectively. The infiltrated HT and MT C/C composites were graphitized under different temperatures of 1600 ℃, 2100 ℃ and 2450℃ for 2h under argon atmosphere. 2.2 Preparation of TEM specimens and characterization methods Specimens were cut into plates with a thickness of 500 μm, and reduced to 80μm thickness by mechanical grinding and double side polishing. Then, the specimens were carefully dimpled down (Dimple Grinder, Gatan 657) to a few microns by using 3 μm and 0.25 μm diamond pastes. Finally, two argon ion guns (Gatan 691) from both sides were operated at 5 kV and a current of 12 mA to specimen surface, and the angle was decreased from 8° to 4°. The duration of the ion thinning process was about 2
Journal Pre-proof
h. At least 3 specimens were carefully prepared for every temperature to observe the interface. The nanostructure and selected area electron diffraction (SAED) of specimens were analyzed by high resolution transmission electron microscopy (HRTEM, FEI Talos F200X & JEM-2100). The orientation angle (OA) can be obtained along the azimuthal intensity distribution of ring-shaped (002) reflections in SAED patterns as depicted by Bourrat et al. [7, 23, 24]. A MATLAB program was used to measure the azimuthal intensity of (002) reflections, then a Gauss fitting was applied to obtain the full width at half maximum (FWHM) of the (002) reflections. The OA was given by the mean value of the FWHM of the
of
ring-shaped (002) reflections.
ro
The morphology of the two C/C composites was observed by polarized light microscopy (PLM, Leica DMLP) and scanning electronic microscopy (SEM, FEI Nova 450). Raman spectrum was
-p
characterized by a Renishaw inVia Raman microscope with a 532 nm laser. Raman analysis was carried out by single spot scanning using a 50× objective lens. The structural parameters were determined by
re
X-ray diffraction (XRD) using a PANalytical X′Pert Pro X-ray diffractometer with Cu Kα radiation.
lP
According to Bragg’s law, the interlayer spacing (d 002) and the height of layered stacking (Lc) were calculated from the (002) diffraction peak using the Bragg equation and the modified Scherrer equation
na
respectively, and the value of the graphitization degree (g) was calculated by Maire and Mering equation [25]. Indentation test was performed by a NANOVEA MHT-M micro-indentation using a Vickers
Jo ur
diamond indenter with a cone angle of 130°, and the load is 5 N. In order to avoid the influence between each indentation during the test, the separation of each indentation is more than five times the distance from the previous indentation. 3 Results
3.1 The PLM and SEM morphologies of HT and MT The polished transverse sections of HT and MT were viewed by PLM, as shown in Fig. 1. The optical activity of MT is not as strong as the HT. There is no obvious microcrack at different temperatures and numerous growth cones can be detected in the HT matrix. However, after treatment at different temperatures, a large number of annular cracks around the carbon fiber were detected clearly in the MT matrix (the yellow arrow area in Fig. 1). The extinction angle (Ae) keeps constant after heat treatment at
Journal Pre-proof
different temperatures [21]. The Ae of HT and MT are 22.22° and 15.74° respectively via the measurement method reported in the study of Li et al. [26], and the corresponding fitted extinction curves are presented in Fig. 2. As mentioned above, on one hand, it proves that the HT and MT C/C composites were prepared successfully. On the other hand, it is found that the microstructure of the two textures has no significant transformation as the temperature increases from PLM observation. The morphologies of HT and MT were further characterized by SEM, as shown in Fig. 3. It can be clearly observed that the cross-section morphology of HT is rough, and there are numerous continuous
of
pyrolytic carbon layers arranged regularly in HT. The surface of MT is relatively smooth, but the carbon
ro
layers are arranged irregularly. The morphology of pyrolytic carbon in some regions of MT is curling and disorder. As the treatment temperature increases from room temperature (around 25 ℃) to 2450 ℃, the
-p
interfacial debonding occurs between matrix and fiber (Fig. 3(g) and 3(h)), and the bonding strength
re
becomes weak. Furthermore, the emergence of nanoscale cracks results from the hydrogen release in pyrolytic carbon at high temperature [27]. The nano-cracks in matrix initiate, propagate and expand to
lP
form microcracks, so the number of microcracks in HT rises sharply. There are multiple bridging lamellae which are used to arrest the propagation of microcracks in the matrix [28], as shown in Fig. 3(c) and 3(e).
na
Compared with HT, the microcrack size in MT matrix is larger, and there is an angle between them (Fig. 3(d) and 3(h)). It is worthy to note that the phenomenon of delamination in MT matrix is different from
Jo ur
that in HT matrix. The HT matrix was delaminated into many nano-thickness sublayers after heat treatment (Fig. 3(c), 3(e), and 3(g)), and the delamination of HT matrix was more apparent as the temperature increases. However, the MT matrix was delaminated into micro-layers including many microcracks and nano-layers (Fig. 3(d), 3(f), and 3(h)). A detailed explanation for the difference will be mentioned later. 3.2 Nanostructure characterization of HT and MT Fig. 4 shows the low-magnification TEM morphologies of HT and MT with different treatment. The interface between the fiber and the matrix can be seen clearly from above images. The interface in the as-deposited state presents a wave shape (Fig. 4(a)) matching the surface of the T300 carbon fiber which is bark-like, rough and has many grooves (Fig. 5(a)). Compared with HT, the microstructure of MT matrix is curlier and more turbulent, as displayed in Fig. 4(a) and 4(b). Moreover, a lot of nanopores can
Journal Pre-proof
be found in MT matrix, and they grow progressively as the temperature rises. The interface width (average of 30 measurement) at different temperatures is given in Fig. 5(b). The interface width of as-deposited HT is about 26.36 nm, and the interface width of HT treated at 1600 °C, 2100 °C and 2450 °C are about 45.43 nm, 62.58 nm and 79.52 nm respectively. The interface width of as-deposited MT is about 34.93 nm, and the interface width of MT treated at 1600 °C, 2100 °C and 2450 °C are approximately 48.97 nm, 70.14 nm and 86.49 nm respectively. Obviously, the interface increases gradually with the temperature rising, and the interface width of HT is always narrower than MT at
of
different temperatures.
ro
Fig. 6 shows the HR-TEM images of HT and MT. It is clearly visible that, both in HT and MT, the carbon layers at the as-deposited state, 1600 °C and 2100 °C are all not as regular as that at 2450 °C. A
-p
series of microcracks and nanopores are detected inside (indicated by yellow arrows). It is important to
re
note that the main defects in MT are nanopores, and numerous nanopores are still observed even after 2450 °C heat treatment (shown in Fig. 6(h)). It can be ascribed to the low orientation of carbon layers in
lP
MT matrix. Although the tortuous carbon layers become ordered during the heat treatment, this process is very slow, and only a small part of carbon layers in MT arrange orderly at 2450 °C. Furthermore, as the
na
temperature increases, nanopores also grow gradually on the nanoscale as the carbon layers become regular [23]. Hence, the nanopores are inevitable. The main defect in HT is microcracks, and the
Jo ur
significant bridging phenomenon is observed in HT, which verified the bridging between nanolayers in the SEM (shown in Fig. 3(c) and 3(e)). And the d002 of the bridging lamellae is about 0.3359 nm, together with its SAED pattern showing that the orientation of bridging lamellae is more ordered than carbon layers in the HT matrix (shown in Fig. 7). 3.3 Micro-indentation test Micro-indentation test was carried on to measure micromechanics of the two C/C composites. The polished HT and MT specimens were tested by micro-indentation with the load of 5 N. Typical depth-load curves of HT and MT with different treatment are shown in Fig. 8(a) and 8(b) respectively. Correspondingly, the micro-hardness and elastic modulus are shown in Fig. 8(c) and 8(d) respectively. As the temperature increasing, the indentation depth of HT and MT becomes deeper, while the indentation depth of HT is deeper than that of MT in four states (shown in dashed line in Fig. 8(a) and 8(b)). The
Journal Pre-proof
micro-hardness of HT and MT decreases after being heat treated, while the decreasing tendency of MT is greater than HT (Fig. 8(c)). Meanwhile, the tendency of elastic modulus is similar to micro-hardness, the elastic modulus of MT is also higher than that of HT after different temperature treatment (Fig. 8(d)). 4 Discussion 4.1 Differences in matrix delamination and interface of HT and MT As can be seen from the SEM results in Fig. 3, the delamination of MT matrix is different from that
of
of HT matrix. HT matrix was directly delaminated into many nano-layers accompanied by multiple microcracks, which could be attributed that the internal defects inside the matrix are reduced and the
ro
graphene layers become flat after heat treatment. In addition, the higher heating treatment temperature of
-p
pyrolytic carbon is, the more orderly the carbon layers arrange, and the more easily they slip [29]. Thus, the number of sublayers rises rapidly. However, MT matrix was delaminated into micro-layers including
re
lots of microcracks and nano-layers. It primarily results from the following factors: (Ⅰ) The higher
lP
thermal expansion coefficient of MT compared to HT(CTE MT>CTEHT) results in high thermal stress on MT [17], so more evident microcracks and interface debonding were captured in MT matrix. (Ⅱ) The
na
emergence of nano cracks resulted from the hydrogen release at high temperature, which can be attributed to the increase in the graphitization (carbon bonding converted from sp 3 to sp2) of pyrolytic carbon [30].
Jo ur
Meanwhile, there were numerous nanopores in MT matrix and they grew progressively as the temperature increased (shown in Fig. 4). Thus, the cracks in MT matrix also might be caused by the change in volume. This may explain why MT matrix suffered more crack than HT matrix. (Ⅲ) Another factor may be due to the differences in nanostructures of various kinds of pyrocarbon [31, 32]. Comparing with MT, the interface width of HT is always narrower at different treatment (Fig. 5(b)). One reasonable factor is that the carbon layers in MT are more curled and stacked than that in HT, and the curled layers on the fiber surface evolve progressively during heat treatment, which needs to occupy more area to complete this evolution. Besides, it is interesting to note that the interface between the carbon fiber and matrix gradually becomes smoother and the span of interface also expands with the increase of treatment temperature (shown in Fig. 4). According to the reports of Zhang et al. [33], the abnormal graphitization near the interface was induced by the difference of thermal expansion between carbon fiber
Journal Pre-proof
and pyrolytic carbon. Therefore, the interface becomes wide and flat during the heat-treatment process. The higher the temperature is, the more obvious above phenomenon will be. 4.2 Lattice parameters transformation of HT and MT
According to the reports of Bourrat [7] and Oberlin [34], the different kinds of pyrocarbon can be described by the size of the structural unit. In order to better characterize the microstructural evolution of
of
HT and MT PyC by semi-quantitating analysis the HRTEM results, a structural unit with preferential
ro
orientation in each HRTEM images was selected to measure its structural data(shown in Fig. 6), including
-p
the length of a straight and perfect fringe (L1) which is close from L a (La is the basal plane length), the
re
thickness of structural unit (L2) which is close from Lc (Lc is the height of layered stacking), and the interlayer spacing between the layers (d002). These structural data were obtained from the average of 50
lP
statistical results. Since the 002 lattice fringes have also the advantage to resolve the problem of a precise
na
localization of the oriented areas [8], the above structural data are yielded directly by 002 lattice fringes.
Jo ur
It has to be mentioned that the data in this work were obtained by statistical data, the value can't accurately reveal the real microcrystalline size, however, they can illustrate the trend of the evolution of the microcrystalline size of pyrocarbon. The structural data are shown in Table. 1.
The OA values are derived from their SAED patterns. The OA value of the original HT is 47 ± 3°, while the OA values of the original MT is 70 ± 3°. The OA values of HT treated at 1600 °C, 2100 °C and 2450 °C are 38 ± 3°, 30 ± 3°, and 25 ± 3° respectively, while the OA values of MT treated at 1600 °C, 2100 °C and 2450 °C are 63 ± 3°, 53 ± 3°, and 50 ± 3°, respectively. It is definitely indicated that the orientation of carbon layers in HT and MT becomes preferable with the increase of treatment temperature. Undoubtedly, the OA values are more convincing to prove the evolution of the carbon layer at the
Journal Pre-proof
micro-level compared to Ae. Meanwhile, it can be found that the d002 of HT decreases from 0.3410 nm to 0.3361 nm, and the d002 of MT decreases from 0.3418 nm to 0.3372 nm. In addition, the value of L2 increases only a little, from 3.86 nm to 5.91 nm in HT and from 3.77 nm to 4.87 nm in MT. The value of L1 however, changes distinctly from 3.40 nm to 10.70 nm in HT and from 3.24 nm to 8.87 nm in MT as shown in Table 1, which means that the increasing tendency of the microcrystalline size of HT is greater
of
than that of MT. In other words, based on the transformation of microcrystalline size, it is distinctly
ro
demonstrated that the size of the microcrystal structure of HT is all larger than that of MT with different
-p
temperature treatment.
XRD and Raman analysis are employed to verify the transformation. XRD patterns and analysis
re
results of HT and MT after different temperature treatment are shown in Fig. 9. It is apparent that all
lP
specimens exhibit a sharp (002) peak, the FWHM of HT is narrower than that of MT, and the 2θ of (002) in two textures shifts to right slightly as the temperature increases, indicating the increase of L c and g of
na
two textures. In addition, the increasing tendency of HT is more distinct than that of MT. Furthermore, by fitting the scatterplot of the angle shift (△θ) from XRD versus variation of interplanar spacing (△d) from
the formula:
Jo ur
HR-TEM (Table 1), as shown in Fig. 10, the variation of △d increases in proportion to △θ and satisfies
𝑦 = 0.61 + 9.34𝑥
(1)
where y is the variation of interplanar spacing value along the fitted curve and x is angle shift. It demonstrates that angle shift reflects directly the crystal size of microstructure. That is to say, the angle shift indirectly indicates the effect of temperature on the evolution of microstructure in C/C composite, which is consistent with the conclusion of HR-TEM. The Raman spectra and analysis results are presented in Fig. 11. There are two obvious vibration peaks in 800-2000 cm-1: the D peak at 1350 cm-1 reflects the defects within the carbon lattice; the G peak at 1580 cm-1 is used to demonstrate the degree of structural order [35-38]. It is not difficult to find that,
Journal Pre-proof
both in HT and MT C/C composites, the FWHM of D peak and G peak becomes narrower as the treatment temperature increases (Fig. 11(a-h)), indicating the increase of crystallinity of two textures. The Raman spectra are fitted with five Lorenz curves located about 1200 cm-1, 1350 cm-1, 1500 cm-1, 1580 cm-1 and 1620 cm-1, respectively [23]. The coefficient R (ID/IG) is obtained by the ratio of the integral intensity of D peak and G peak to characterize graphitization degree [39], and the La value has an inverse proportion relationship with R value, i.e., La=4.4R-1. The results of R (ID/IG) and La are shown in Fig. 11 (i) and (j) respectively. It can be seen that the decrease of R and the increase of L a of HT after heat treatment
of
are more obvious than those of MT, indicating higher graphitization of HT compared to MT at every temperature. The above phenomenon presents that the type of carbon bonding in HT and MT has
ro
transformed from sp3 to sp2 as the stacking method in carbon atoms changes. Meanwhile, from the
than that of MT after different heat treatment.
-p
changing tendency of the R value, it is further explained that the structural order degree of HT is greater
re
4.3 The influence of microstructural evolution on micro-hardness
lP
Generally, high temperature provides driving energy for rearrangement of carbon atoms, leading to the tortuous carbon eventually transformed into the graphite. Oberlin et al. [40] introduced that the
na
turbostratic structure started to disappear at a temperature range of 1500 °C-1700 °C and formed distorted layers. Hence, only a few of carbon atoms in the turbulence structure were rearranged at 1600 °C. The
Jo ur
rearrangement of carbon atoms further enhanced above 2100 °C, and more carbon atoms were rearranged regularly up to 2450 °C, as shown in Fig. 12(a). Surely, the process of atomic rearrangement corresponds to the transformation of microstructure in pyrolytic carbon, and the pyrolytic carbon in the as-deposited state contains a large amount of C-C bonds (σ-bonds) with a bond angle of 109.5°. It is quite easy to form defects, thus, the carbon layers are disorderly arranged in the as-deposited state. And the hydrogen release in pyrolytic carbon occurs at high temperature [27]. The C-C single bonds decrease, while C=C double bonds increase with the increasing temperature shown in Fig. 12(b). The shape of C=C double bonds is similar to a benzene ring whose carbon atoms are in the same plane, and this bond is also called conjugate big π-bond with a stable structure, which is beneficial to the orientation of carbon atoms. Moreover, because the HT includes more C=C bonds than MT, the carbon layers of HT are more ordered than that in MT in the as- deposited state. Therefore, the orientation of the carbon layer in HT is greater than MT at all heat-treatment temperature. It is consistent to the observation of HRTEM images shown in Fig. 6.
Journal Pre-proof
Since the microstructures of PyC is the main factor influencing the mechanical properties of C/C composites, the mechanical properties vary accompanied by the microstructural evolution. There is no exception of micro-hardness and modulus for HT and MT. According to the report of Lieberman et al [41], the HT matrix was relatively soft compared with MT matrix attributed to the difference in internal structure. As shown in Fig. 13, since the carbon layers in the MT are curled and stacked, the carbon layers in MT are tightly entangled with each other resulting in strong resistance when the indenter is pressed into the matrix. The carbon layers in HT are better aligned, which is good for slipping. Therefore, the carbon
of
layers are compressed directly when the force is applied. Hence, the resistance to external forces of MT is greater than that of HT, and the micro-hardness of the MT is larger than HT. In addition, the elastic
ro
modulus of MT is also higher than that of HT because of the relatively small deformation. Based on the
re
and HT specimens after heat treatment.
-p
microstructural evolution of HT and MT mentioned above, this conclusion is also applicable to the MT
5 Conclusions
lP
1. After treatment at different temperatures, the indentation depth of HT is always deeper than that of
na
MT, and the corresponding micro-hardness and elastic modulus of MT are also greater than that of HT. 2. The delamination of MT and HT was different after heat treatment. HT matrix was directly
Jo ur
delaminated into many nano-layers accompanied by multiple microcracks. However, MT matrix was delaminated into micro-layers in which there were lots of microcracks and nano-layers. 2. The interface width of HT is always narrower than that of MT, and there are numerous nanopores in MT even after heat treatment at 2450 °C. 4. The bridging phenomenon is markedly observed in HT and the d 002 of the bridging lamella with a preferable orientation is about 0.3359 nm. 5. A fitted formula is proposed to describe the relationship between the variation of interplanar spacing from HR-TEM and the angle shift from XRD, indicating that interplanar spacing is in proportion to angle shift. Declaration of Interest
Journal Pre-proof
There are no interests to declare. Acknowledgments We would like to thank the Analytical & Testing Center of Northwestern Polytechnical University for providing TEM technical support. The financial supported by the National Natural Science Foundation of China under Grant Nos.51821091, 51432008. References
of
[1] E. Fitzer. The future of carbon-carbon composites. Carbon, 1987, 25(2): 163-190.
ro
[2] J.E. Sheehan, K.W. Buesking, B.J. Sullivan. Carbon-carbon composites. Annual Review of Materials Science, 1994, 24(1): 19-44.
-p
[3] H.J. Li, A.J. Li, R.C. Bai, K.Z. Li. Numerical simulation of chemical vapor infiltration of propylene
re
into C/C composites with reduced multi-step kinetic models. Carbon, 2005, 43(14): 2937-2950. [4] Q.L. Shen, Q. Song, H.J. Li, C.X. Xiao, T.Y. Wang, H.J. Lin, W. Li. Fatigue strengthening of
lP
carbon/carbon composites modified with carbon nanotubes and silicon carbide nanowires. International Journal of Fatigue, 2019, 124: 411-421.
na
[5] Q. Song, K.Z. Li, H.J. Li, S.Y. Zhang, L.H. Qi, Q.G. Fu. A novel method to fabricate isotropic pyrocarbon by densifying a multi-walled carbon nanotube preform by fixed-bed chemical vapor
Jo ur
deposition. Carbon, 2013, 59: 547-550.
[6] R.J. Gray, J.V. Cathcart. Polarized light microscopy of pyrolytic carbon deposits. Journal of Nuclear Materials, 1966, 19(1): 81-89.
[7] X. Bourrat, B. Trouvat, G. Limousin, G. Vignoles, F. Doux. Pyrocarbon anisotropy as measured by electron diffraction and polarized light. Journal of Materials Research, 2000, 15(1): 92-101. [8] B. Reznik, K.J. Hüttinger. On the terminology for pyrolytic carbon. Carbon, 2002, 40(4): 621-624. [9] H.C. Cao, E. Bischoff, O. Sbaizero, M. Rühle, A.G. Evans, D.B. Marshall, J.J. Brennan. Effect of interfaces on the properties of fiber‐ reinforced ceramics. Journal of the American Ceramic Society, 1990, 73(6): 1691-1699. [10] T.P. Weihs, O. Sbaizero, E.Y. Luh, W.D. Nix. Correlating the mechanical properties of a continuous fiber‐ reinforced ceramic‐ matrix composite to the sliding resistance of the fibers. Journal of the
Journal Pre-proof
American Ceramic Society, 1991, 74(3): 535-540. [11] S.J. Park, M.H. Kim. Effect of acidic anode treatment on carbon fibers for increasing fiber-matrix adhesion and its relationship to interlaminar shear strength of composites. Journal of Materials Science, 2000, 35(8): 1901-1905. [12] F.H. Zhang, R.G. Wang, X.D. He, C. Wang, L.N. Ren. Interfacial shearing strength and reinforcing mechanisms of an epoxy composite reinforced using a carbon nanotube/carbon fiber hybrid. Journal of Materials Science, 2009, 44(13): 3574-3577.
of
[13] Q. Song, K.Z. Li, L.H. Qi, H.J. Li, J.H. Lu, L.L. Zhang, Q.G. Fu. The reinforcement and toughening
ro
of pyrocarbon-based carbon/carbon composite by controlling carbon nanotube growth position in carbon felt. Materials Science & Engineering A Structural Materials Properties Microstructure &
-p
Processing, 2013, 564: 71-75.
re
[14] X.J. Chao, L.H. Qi, W.L. Tian, X.H. Hou, W.J. Ma, H.J. Li. Numerical evaluation of the influence of porosity on bending properties of 2D carbon/carbon composites. Composites Part B: Engineering,
lP
2018, 136: 72-80.
1998.
na
[15] E. Fitzer, L.M. Manocha, Carbon reinforcements and carbon/carbon composites. Springer, Berlin,
[16] W.G. Zhang, Z.J. Hu, K.J. Hüttinger. Chemical vapor infiltration of carbon fiber felt: optimization of
Jo ur
densification and carbon microstructure. Carbon, 2002, 40(14): 2529-2545. [17] H.O. Pierson, D.A. Northrop. Carbon-felt, carbon-matrix composites: dependence of thermal and mechanical properties on fiber precursor and matrix structure. Journal of Composite Materials 1974, 9: 118-137.
[18] R. Piat, Y. Lapusta, T. Böhlke, M. Guellali, B. Reznik, D. Gerthsen, T.F. Chen, R. Oberacker, M.J. Hoffmann. Microstructure-induced thermal stresses in pyrolytic carbon matrices at temperatures up to 2900° C. Journal of the European Ceramic Society, 2007, 27(16): 4813-4820. [19] S. Ozcan, J. Tezcan, B. Gurung, P. Filip. The effect of heat treatment temperature on the interfacial shear strength of C/C composites. Journal of Materials Science, 2010, 46(1): 38-46. [20] K. Fujita, H. Sakai, N. Iwashita, Y. Sawada. Influence of heat treatment temperature on interfacial shear strength of C/C. Composites Part A: Applied Science and Manufacturing, 1999, 30(4):
Journal Pre-proof
497-501. [21] L.H. Xia, B.Y. Huang, F.Q. Zhang, Z.M. Liu, T.F. Chen. Effect of heat treatment on cracking and strength of carbon/carbon composites with smooth laminar pyrocarbon matrix. Materials & Design, 2016, 107: 33-40. [22] W. Li, H.J. Li, J. Wang, S.Y. Zhang, X. Yang, J.F. Wei. Preparation and mechanical properties of carbon/carbon composites with high textured pyrolytic carbon matrix. Transactions of Nonferrous Metals Society of China, 2013, 23(7): 2129-2134.
of
[23] G.H. Zhou, Y.Q. Liu, L.L. He, Q.H. Guo, H.Q. Ye. Microstructure difference between core and skin
ro
of T700 carbon fibers in heat-treated carbon/carbon composites. Carbon, 2011, 49(9): 2883-2892. [24] A. Tressaud, M. Chambon, V. Gupta, S. Flandrois, O.P. Bahl. Fluorine-intercalated carbon fibers III.
-p
A transmission electron microscopy study. Carbon, 1995, 33(9): 1339-1345.
re
[25] W. Yang, R.Y. Luo, Z.H. Hou, Y. Zhang, H.D. Shang, M.Y. Hao. Influence of the microstructure of the carbon matrices on the internal friction behavior of carbon/carbon composites. New Carbon
lP
Materials, 2016, 31(2): 159-166.
[26] M.L. Li, L.H. Qi, H.J. Li, G.Z. Xu. Measurement of the extinction angle about laminar pyrocarbons
na
by image analysis in reflection polarized light. Materials Science and Engineering: A, 2007, 448(1-2): 80-87.
Jo ur
[27] B. Reznik, K. Norinaga, D. Gerthsen, O. Deutschmann. The effect of cooling rate on hydrogen release from a pyrolytic carbon coating and its resulting morphology. Carbon, 2006, 44(7): 1330-1334.
[28] B. Reznik, M. Fotouhi. Electron microscopy and electron-energy-loss spectroscopy study of crack bridging in carbon–carbon composites. Composites Science & Technology, 2008, 68(5): 1131-1135. [29] J.J. Ren, K.Z. Li, S.Y. Zhang, X.Y. Yao, H.J. Li. Preparation of high texture three-dimensional braided carbon/carbon composites by pyrolysis of ethanol and methane. Ceramics International, 2016, 42(2): 2887-2891. [30] F. Cancino-Trejo, M. Sáenz Padilla, E. López-Honorato, U. Carvajal-Nunez, J. Boshoven, J. Somers. The effect of heat treatment on the microstructure and diffusion of silver in pyrolytic carbon coatings. Carbon, 2016, 109: 542-551.
Journal Pre-proof
[31] A. Pfrang, B. Reznik, D. Gerthsen, T. Schimmel. Comparative study of differently textured pyrolytic carbon layers by atomic force, transmission electron and polarized light microscopy. Carbon, 2003, 41(1): 181-185. [32] B. Reznik, D. Gerthsen, K.J. Hüttinger. Micro- and nanostructure of the carbon matrix of infiltrated carbon fiber felts. Carbon, 2001, 39(2): 215-229. [33] F.Q. Zhang, B.Y. Huang, Q.Z. Huang, X. Xiong, T.F. Cheng. Effects of the interface on the graphitization of a carbon fiber/pyrolytic carbon composite. Carbon, 2003, 41(3): 610-612.
of
[34] A. Oberlin. Pyrocarbons. Carbon, 2002, 40(1): 7-24.
graphene. Applied Physics Letters, 2008, 92(4): 666.
ro
[35] Y.Y. Wang, Z.H. Ni, Z.X. Shen, H.M. Wang, Y.H. Wu. Interference enhancement of Raman signal of
-p
[36] D. Chotikapanich, W.E. Griffiths. Estimating Lorenz curves using a Dirichlet distribution. Journal of
re
Business & Economic Statistics, 2002, 20(2): 290-295.
[37] Q. Song, H.B. Yan, K.D. Liu, K.Y. Xie, W. Li, W.H. Gai, G.H. Chen, H.J. Li, C. Shen, Q.G. Fu.
lP
Vertically grown edge‐ rich graphene nanosheets for spatial control of Li nucleation. Advanced Energy Materials, 2018, 8(22): 1800564.
Lightweight
and
na
[38] X.M. Yin, H.J. Li, L.Y. Han, J.C. Meng, J.H. Lu, L.L. Zhang, W. Li, Q.G. Fu, K.Z. Li, Q. Song. flexible
graphene
shielding.
Jo ur
electromagnetic-interference
3D
microtubes Chemical
membrane Engineering
for
high-efficiency
Journal,
2020:
10.1016/j.cej.2020.124025.
[39] C. Casiraghi, S. Pisana, K.S. Novoselov, A.K. Geim, A.C. Ferrari. Raman fingerprint of charged impurities in graphene. Applied Physics Letters, 2007, 91(23): 183. [40] A. Oberlin. carbonization and graphitization. Carbon, 1984, 22(6): 521-541. [41] M.L. Lieberman, H.O. Pierson. Effect of gas phase conditions on resultant matrix pyrocarbons in carbon/carbon composites. Carbon, 1974, 12(3): 233-241.
Journal Pre-proof
Declaration of interests The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Jo ur
na
lP
re
-p
ro
of
☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:
Journal Pre-proof
Credit Author Statement
Jo ur
na
lP
re
-p
ro
of
Tiyuan Wang: Methodology, software, investigation, formal writing-original draft preparation; Hejun Li: Conceptualization, resources and funding acquisition; Shouyang Zhang: writing-review and editing, Conceptualization; Kun Li: Visualization, supervision and project administration; Wei Li: Data curation, Validation.
analysis
and
Journal Pre-proof
of
Fig. 1 Polarized light micrographs of HT(a-d) and MT(e-h) with different treatment: (a, e)as-deposited
Jo ur
na
lP
re
-p
ro
state; (b, f)1600°C; (c, g)2100°C; (d, h)2450°C
of
Journal Pre-proof
Jo ur
na
lP
re
-p
ro
Fig. 2 The fitted extinction curves and the corresponding extinction angle of HT(a) and MT(b).
Jo ur
na
lP
re
-p
ro
of
Journal Pre-proof
Fig. 3 SEM images of HT(a, c, e, g) and MT(b, d, f, h) with different treatment: (a, b)as-deposited state; (c, d)1600°C; (e, f)2100°C; (g, h)2450°C.
Jo ur
na
lP
re
-p
ro
of
Journal Pre-proof
Journal Pre-proof
Fig. 4 TEM images of HT(a, c, e, g) and MT(b, d, f, h) with different treatment: (a, b)as-deposited state;
-p
ro
of
(c, d)1600°C; (e, f)2100°C; (g, h)2450°C.
Figure.5 The surface morphology of carbon fiber (a) and the interface width of HT and MT under four
Jo ur
na
lP
re
temperatures (b).
Jo ur
na
lP
re
-p
ro
of
Journal Pre-proof
Jo ur
na
lP
re
-p
ro
of
Journal Pre-proof
Fig. 6 HR-TEM micrographs of HT(a, c, e, g) and MT(b, d, f, h) with different treatment: (a, b)as-deposited state; (c, d)1600°C; (e, f)2100°C; (g, h)2450°C.
Journal Pre-proof
Fig. 7 The bridging image in HT:(a)low-magnification image;(b)overview of microcrack; (c)HR-TEM
Jo ur
na
lP
re
-p
ro
of
image of (b).
lP
re
-p
ro
of
Journal Pre-proof
na
Fig. 8 Micro-indentation curves and results of two textures with different treatment: (a)micro-indentation
Jo ur
curves of HT; (b)micro-indentation curves of MT; (c)micro-hardness results; (d)elastic modulus results.
lP
re
-p
ro
of
Journal Pre-proof
Fig. 9 XRD patterns and analysis results of HT and MT with different treatment: (a) XRD patterns of HT;
Jo ur
na
(b)XRD patterns of MT; (c) Lc results of HT and MT; (d) graphitization degree of HT and MT.
re
-p
ro
of
Journal Pre-proof
Jo ur
na
lP
Fig. 10 Plot of angle shift (△θ) from XRD versus variation of interplanar spacing (△d) from HR-TEM
Jo ur
na
lP
re
-p
ro
of
Journal Pre-proof
Fig. 11 Raman spectra of HT(a, c, e, g) and MT(b, d, f, h) with different treatment: (a, b)as-deposited state;
Journal Pre-proof
lP
re
-p
ro
of
(c, d)1600°C; (e, f)2100°C; (g, h)2450°C; and (i)R(ID/IG) results; (j) La results.
Fig. 12 Sketches of the different microstructures of MT and HT (a) and the transformation of bonds
Jo ur
na
before and after HTT(b).
lP
re
-p
ro
of
Journal Pre-proof
Jo ur
na
Fig. 13 Schematic diagram of carbon layers change before and after indentation.
Journal Pre-proof
Table.1 Statistical structural data obtained from HR-TEM images. L1/(nm)
L2/(nm)
OA/°
HT
0.3410±0.001
3.40±0.04
3.86±0.05
47±3°
HT-1600°C
0.3396±0.002
4.96±0.05
4.14±0.03
38±3°
HT-2100°C
0.3365±0.001
6.98±0.003
4.49±0.04
30±3°
HT-2450°C
0.3361±0.001
10.70±0.05
5.91±0.04
25±3°
MT
0.3418±0.002
3.24±0.04
3.77±0.05
70±3°
MT-1600°C
0.3404±0.001
4.78±0.05
3.88±0.04
64±3°
MT-2100°C
0.3387±0.001
5.94±0.04
4.11±0.03
53±3°
MT-2450°C
0.3372±0.002
8.87±0.05
4.87±0.05
50±3°
Jo ur
na
lP
re
-p
ro
of
d002/(nm)
Journal Pre-proof
of
Graphical abstract
Jo ur
na
lP
re
-p
ro
Graphical abstract
Journal Pre-proof
Highlights 1.In-depth observation of microstructural evolution of different pyrolytic carbon
2.The microstructural distinctions between different pyrocarbon were compared at multiple scales
3.The influence of microstructural evolution on the micromechanics of composites was
Jo ur
na
lP
re
-p
ro
of
analyzed
Tiyuan Wang, Hejun Li, Shouyang Zhang, Kun Li, Wei Li PII:
S0925-9635(19)30818-0
DOI:
https://doi.org/10.1016/j.diamond.2020.107729
Reference:
DIAMAT 107729
To appear in:
Diamond & Related Materials
Received date:
22 November 2019
Revised date:
9 January 2020
Accepted date:
22 January 2020
Please cite this article as: T. Wang, H. Li, S. Zhang, et al., The effect of microstructural evolution on micromechanical behavior of pyrolytic carbon after heat treatment, Diamond & Related Materials (2020), https://doi.org/10.1016/j.diamond.2020.107729
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. 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.
© 2020 Published by Elsevier.
Journal Pre-proof
Title page
Title: The effect of microstructural evolution on micromechanical behavior of pyrolytic carbon after heat treatment
Author names and affiliations: Tiyuan Wang, Hejun Li, Shouyang Zhang, Kun Li,
of
Wei Li (State Key Laboratory of Solidification Processing, Northwestern Polytechnical
ro
University, Xi’an, Shaanxi 710072, PR China)
-p
Corresponding author: Hejun Li (Tel.: +86-29-88495004. E-mail address:
lP
Jo ur
na
Declarations of interest: none
re
[email protected])
Journal Pre-proof
The effect of microstructural evolution on micromechanical behavior of pyrolytic carbon after heat treatment Tiyuan Wang, Hejun Li*, Shouyang Zhang*, Kun Li, Wei Li State Key Laboratory of Solidification Processing, Northwestern Polytechnical University, Xi’an, Shaanxi 710072, PR China Abstract
of
In order to investigate the effect of microstructural evolution on the micromechanical behavior of
ro
pyrolytic carbon (PyC), the microstructural parameters and evolution of high texture (HT) and medium texture (MT) PyC after heat treatment from 1600 °C to 2450 °C were studied comparatively. The
-p
micro-indentation test was also performed to reveal the micromechanics. In terms of microstructure, after
re
heat treatment, the HT matrix was directly delaminated into many nano-layers accompanied by multiple microcracks, while the MT matrix was delaminated into micro-layers including lots of cracks. Meanwhile,
lP
numerous nanopores were still observed in the MT matrix after heat treatment at 2450 °C. The interfaces of two textures became smoother and wider, and the width of HT interface was narrower than that of MT
na
at different heat treatment temperatures. Moreover, the increasing tendency of the basal plane length (La) and the height of layered stacking (Lc) of HT were greater than those of MT as HT PyC contained more
Jo ur
C=C bonds than MT PyC. The micro-indentation results showed that the indentation depth of HT was always deeper than that of MT, and the corresponding micro-hardness and elastic modulus of MT were also greater than that of HT after different temperature treatment. Key words: Pyrolytic carbon; Microstructure; Micro-hardness; Lattice structure; Interface 1 Introduction Carbon/Carbon (C/C) composites exhibit superior high temperature performance including low density, high specific strength, high fracture toughness, and outstanding thermal shock resistance, which makes them widely used as throat linings, aircraft brake discs, and nose cones [1-5]. Chemical vapor
*
Corresponding author. Tel.: +86 029 88495004; E-mail address: [email protected] (H. Li)
*
Corresponding author. E-mail addresses: [email protected] (S. Zhang).
Journal Pre-proof
infiltration (CVI) is the most common process for fabricating C/C composites. In the view of microstructures, the first classification of different microstructures of pyrolytic carbon was presented by Gray and Cathcart [6]. Bourrat et al. [7] further classified the microstructure of pyrolytic carbon using extinction angle (Ae) of polarized light microscope (PLM) and selected area electron diffraction (SAED) of TEM observation. Up to now, the pyrolytic carbon usually has four typical textures: high texture (HT), medium texture (MT), low texture (LT) and isotropic carbon (ISO), which was proposed by Reznik and Hüttinger [8]. The mechanical properties of C/C composites depend on the microstructures of PyC, the
of
type of carbon fiber and the state of interface bonding between fiber and matrix [9-14]. On the other hand, graphitization, as a post-treatment after the infiltration process, is an essential factor for the mechanics of
ro
C/C composites [15, 16]. So far, numerous studies concentrate on the effect of heat treatment on
-p
mechanical properties, but the theoretical contribution on microstructures in this field is far from enough.
re
The heat treatment temperature has an essential influence on the microstructure and mechanical properties of C/C material. In the early 1970s, Pierson et al. [17] found that the flexural strength and
lP
modulus of C/C composites decreased after high heat treatment temperature (HTT). Piat et al. [18] reported the effect of heat treatment temperature on the mechanical properties of tangential fiber
na
reinforced MT and HT. The results showed that with the increase of heat treatment temperature (2200-2900 °C), the strength of HT material decreased and exhibited typical pseudoplasticity, while the
Jo ur
strength of MT gradually increased and presented obvious brittleness. Ozcan et al. [19] investigated the effect of HTT on interfacial shear strength (IFSS) of C/C composites and detected that compared to heat treatment at 1800 °C, the IFSS of C/C composites decreased after heat treatment at 2100 °C and 2400 °C. This result was in good agreement with another investigation [20], and it can be partially attributed to the reorganization of the graphene sheets in the matrix in the vicinity of fiber/matrix interface. Xia et al. [21] analyzed the microstructure and flexural properties of C/C composites after heat treatment from 1800 °C to 2500 °C. It was found that heat treatment resulted in distinct interfacial cracks and concentric cracks in smooth laminar pyrolytic carbon and the flexural strength decreased rapidly after treated at 2100 °C while increased back at 2450 °C. The reason was that the poor stress transfer capabilities and high stress concentration caused by multiple interfacial debonding and concentric cracks. Li et al. [22] reported that the crystalline thickness, graphitization degree, flexural strength and modulus of C/C composites with HT matrix increased obviously after heat treatment at 2500 °C.
Journal Pre-proof
For a long time, there is a lack of detailed and sufficient analyses about the effect of microstructure evolution on micromechanics of pyrolytic carbon after heat treatment, even though the mechanical properties of C/C composites were probed by many above researches. Thus, a further characterization of the structure of C/C composites in both micro and nano scale is necessary owing to the insufficient correlation between the microstructure and mechanical properties. Moreover, a thorough understanding of their microstructural evolution and lattice parameters after heat treatment is of great significance for the design, application and performance evaluation of the C/C materials. Hence, in this work, the
of
micro-hardness was tested to evaluate the micromechanics of HT and MT under different treatment temperatures, comprehensively considering the delamination, interface, lattice structure morphologies.
ro
Meanwhile, microstructure evolution and bond transformation were also used to explain the difference in
-p
micro-hardness.
re
2 Material and Experiment
lP
2.1 Specimens preparation
2.5D needled PAN-based carbon fiber felts (Yixing Tianniao Co., Ltd., China) with a density of 0.45
na
g/cm3 were used as preforms. Thermal gradient chemical vapor infiltration was employed to infiltrate the preforms at the temperature range of 900 °C ~ 1200 °C, using natural gas (>95%) and argon as the
Jo ur
precursor and carrier gas. The HT and MT C/C composites were finally obtained by controlling the process parameters. The bulk density and porosity of HT and MT C/C composites measured by the Archimedes method were1.78 g/cm3, 8.3%, and 1.75 g/cm3, 8.6% respectively. The infiltrated HT and MT C/C composites were graphitized under different temperatures of 1600 ℃, 2100 ℃ and 2450℃ for 2h under argon atmosphere. 2.2 Preparation of TEM specimens and characterization methods Specimens were cut into plates with a thickness of 500 μm, and reduced to 80μm thickness by mechanical grinding and double side polishing. Then, the specimens were carefully dimpled down (Dimple Grinder, Gatan 657) to a few microns by using 3 μm and 0.25 μm diamond pastes. Finally, two argon ion guns (Gatan 691) from both sides were operated at 5 kV and a current of 12 mA to specimen surface, and the angle was decreased from 8° to 4°. The duration of the ion thinning process was about 2
Journal Pre-proof
h. At least 3 specimens were carefully prepared for every temperature to observe the interface. The nanostructure and selected area electron diffraction (SAED) of specimens were analyzed by high resolution transmission electron microscopy (HRTEM, FEI Talos F200X & JEM-2100). The orientation angle (OA) can be obtained along the azimuthal intensity distribution of ring-shaped (002) reflections in SAED patterns as depicted by Bourrat et al. [7, 23, 24]. A MATLAB program was used to measure the azimuthal intensity of (002) reflections, then a Gauss fitting was applied to obtain the full width at half maximum (FWHM) of the (002) reflections. The OA was given by the mean value of the FWHM of the
of
ring-shaped (002) reflections.
ro
The morphology of the two C/C composites was observed by polarized light microscopy (PLM, Leica DMLP) and scanning electronic microscopy (SEM, FEI Nova 450). Raman spectrum was
-p
characterized by a Renishaw inVia Raman microscope with a 532 nm laser. Raman analysis was carried out by single spot scanning using a 50× objective lens. The structural parameters were determined by
re
X-ray diffraction (XRD) using a PANalytical X′Pert Pro X-ray diffractometer with Cu Kα radiation.
lP
According to Bragg’s law, the interlayer spacing (d 002) and the height of layered stacking (Lc) were calculated from the (002) diffraction peak using the Bragg equation and the modified Scherrer equation
na
respectively, and the value of the graphitization degree (g) was calculated by Maire and Mering equation [25]. Indentation test was performed by a NANOVEA MHT-M micro-indentation using a Vickers
Jo ur
diamond indenter with a cone angle of 130°, and the load is 5 N. In order to avoid the influence between each indentation during the test, the separation of each indentation is more than five times the distance from the previous indentation. 3 Results
3.1 The PLM and SEM morphologies of HT and MT The polished transverse sections of HT and MT were viewed by PLM, as shown in Fig. 1. The optical activity of MT is not as strong as the HT. There is no obvious microcrack at different temperatures and numerous growth cones can be detected in the HT matrix. However, after treatment at different temperatures, a large number of annular cracks around the carbon fiber were detected clearly in the MT matrix (the yellow arrow area in Fig. 1). The extinction angle (Ae) keeps constant after heat treatment at
Journal Pre-proof
different temperatures [21]. The Ae of HT and MT are 22.22° and 15.74° respectively via the measurement method reported in the study of Li et al. [26], and the corresponding fitted extinction curves are presented in Fig. 2. As mentioned above, on one hand, it proves that the HT and MT C/C composites were prepared successfully. On the other hand, it is found that the microstructure of the two textures has no significant transformation as the temperature increases from PLM observation. The morphologies of HT and MT were further characterized by SEM, as shown in Fig. 3. It can be clearly observed that the cross-section morphology of HT is rough, and there are numerous continuous
of
pyrolytic carbon layers arranged regularly in HT. The surface of MT is relatively smooth, but the carbon
ro
layers are arranged irregularly. The morphology of pyrolytic carbon in some regions of MT is curling and disorder. As the treatment temperature increases from room temperature (around 25 ℃) to 2450 ℃, the
-p
interfacial debonding occurs between matrix and fiber (Fig. 3(g) and 3(h)), and the bonding strength
re
becomes weak. Furthermore, the emergence of nanoscale cracks results from the hydrogen release in pyrolytic carbon at high temperature [27]. The nano-cracks in matrix initiate, propagate and expand to
lP
form microcracks, so the number of microcracks in HT rises sharply. There are multiple bridging lamellae which are used to arrest the propagation of microcracks in the matrix [28], as shown in Fig. 3(c) and 3(e).
na
Compared with HT, the microcrack size in MT matrix is larger, and there is an angle between them (Fig. 3(d) and 3(h)). It is worthy to note that the phenomenon of delamination in MT matrix is different from
Jo ur
that in HT matrix. The HT matrix was delaminated into many nano-thickness sublayers after heat treatment (Fig. 3(c), 3(e), and 3(g)), and the delamination of HT matrix was more apparent as the temperature increases. However, the MT matrix was delaminated into micro-layers including many microcracks and nano-layers (Fig. 3(d), 3(f), and 3(h)). A detailed explanation for the difference will be mentioned later. 3.2 Nanostructure characterization of HT and MT Fig. 4 shows the low-magnification TEM morphologies of HT and MT with different treatment. The interface between the fiber and the matrix can be seen clearly from above images. The interface in the as-deposited state presents a wave shape (Fig. 4(a)) matching the surface of the T300 carbon fiber which is bark-like, rough and has many grooves (Fig. 5(a)). Compared with HT, the microstructure of MT matrix is curlier and more turbulent, as displayed in Fig. 4(a) and 4(b). Moreover, a lot of nanopores can
Journal Pre-proof
be found in MT matrix, and they grow progressively as the temperature rises. The interface width (average of 30 measurement) at different temperatures is given in Fig. 5(b). The interface width of as-deposited HT is about 26.36 nm, and the interface width of HT treated at 1600 °C, 2100 °C and 2450 °C are about 45.43 nm, 62.58 nm and 79.52 nm respectively. The interface width of as-deposited MT is about 34.93 nm, and the interface width of MT treated at 1600 °C, 2100 °C and 2450 °C are approximately 48.97 nm, 70.14 nm and 86.49 nm respectively. Obviously, the interface increases gradually with the temperature rising, and the interface width of HT is always narrower than MT at
of
different temperatures.
ro
Fig. 6 shows the HR-TEM images of HT and MT. It is clearly visible that, both in HT and MT, the carbon layers at the as-deposited state, 1600 °C and 2100 °C are all not as regular as that at 2450 °C. A
-p
series of microcracks and nanopores are detected inside (indicated by yellow arrows). It is important to
re
note that the main defects in MT are nanopores, and numerous nanopores are still observed even after 2450 °C heat treatment (shown in Fig. 6(h)). It can be ascribed to the low orientation of carbon layers in
lP
MT matrix. Although the tortuous carbon layers become ordered during the heat treatment, this process is very slow, and only a small part of carbon layers in MT arrange orderly at 2450 °C. Furthermore, as the
na
temperature increases, nanopores also grow gradually on the nanoscale as the carbon layers become regular [23]. Hence, the nanopores are inevitable. The main defect in HT is microcracks, and the
Jo ur
significant bridging phenomenon is observed in HT, which verified the bridging between nanolayers in the SEM (shown in Fig. 3(c) and 3(e)). And the d002 of the bridging lamellae is about 0.3359 nm, together with its SAED pattern showing that the orientation of bridging lamellae is more ordered than carbon layers in the HT matrix (shown in Fig. 7). 3.3 Micro-indentation test Micro-indentation test was carried on to measure micromechanics of the two C/C composites. The polished HT and MT specimens were tested by micro-indentation with the load of 5 N. Typical depth-load curves of HT and MT with different treatment are shown in Fig. 8(a) and 8(b) respectively. Correspondingly, the micro-hardness and elastic modulus are shown in Fig. 8(c) and 8(d) respectively. As the temperature increasing, the indentation depth of HT and MT becomes deeper, while the indentation depth of HT is deeper than that of MT in four states (shown in dashed line in Fig. 8(a) and 8(b)). The
Journal Pre-proof
micro-hardness of HT and MT decreases after being heat treated, while the decreasing tendency of MT is greater than HT (Fig. 8(c)). Meanwhile, the tendency of elastic modulus is similar to micro-hardness, the elastic modulus of MT is also higher than that of HT after different temperature treatment (Fig. 8(d)). 4 Discussion 4.1 Differences in matrix delamination and interface of HT and MT As can be seen from the SEM results in Fig. 3, the delamination of MT matrix is different from that
of
of HT matrix. HT matrix was directly delaminated into many nano-layers accompanied by multiple microcracks, which could be attributed that the internal defects inside the matrix are reduced and the
ro
graphene layers become flat after heat treatment. In addition, the higher heating treatment temperature of
-p
pyrolytic carbon is, the more orderly the carbon layers arrange, and the more easily they slip [29]. Thus, the number of sublayers rises rapidly. However, MT matrix was delaminated into micro-layers including
re
lots of microcracks and nano-layers. It primarily results from the following factors: (Ⅰ) The higher
lP
thermal expansion coefficient of MT compared to HT(CTE MT>CTEHT) results in high thermal stress on MT [17], so more evident microcracks and interface debonding were captured in MT matrix. (Ⅱ) The
na
emergence of nano cracks resulted from the hydrogen release at high temperature, which can be attributed to the increase in the graphitization (carbon bonding converted from sp 3 to sp2) of pyrolytic carbon [30].
Jo ur
Meanwhile, there were numerous nanopores in MT matrix and they grew progressively as the temperature increased (shown in Fig. 4). Thus, the cracks in MT matrix also might be caused by the change in volume. This may explain why MT matrix suffered more crack than HT matrix. (Ⅲ) Another factor may be due to the differences in nanostructures of various kinds of pyrocarbon [31, 32]. Comparing with MT, the interface width of HT is always narrower at different treatment (Fig. 5(b)). One reasonable factor is that the carbon layers in MT are more curled and stacked than that in HT, and the curled layers on the fiber surface evolve progressively during heat treatment, which needs to occupy more area to complete this evolution. Besides, it is interesting to note that the interface between the carbon fiber and matrix gradually becomes smoother and the span of interface also expands with the increase of treatment temperature (shown in Fig. 4). According to the reports of Zhang et al. [33], the abnormal graphitization near the interface was induced by the difference of thermal expansion between carbon fiber
Journal Pre-proof
and pyrolytic carbon. Therefore, the interface becomes wide and flat during the heat-treatment process. The higher the temperature is, the more obvious above phenomenon will be. 4.2 Lattice parameters transformation of HT and MT
According to the reports of Bourrat [7] and Oberlin [34], the different kinds of pyrocarbon can be described by the size of the structural unit. In order to better characterize the microstructural evolution of
of
HT and MT PyC by semi-quantitating analysis the HRTEM results, a structural unit with preferential
ro
orientation in each HRTEM images was selected to measure its structural data(shown in Fig. 6), including
-p
the length of a straight and perfect fringe (L1) which is close from L a (La is the basal plane length), the
re
thickness of structural unit (L2) which is close from Lc (Lc is the height of layered stacking), and the interlayer spacing between the layers (d002). These structural data were obtained from the average of 50
lP
statistical results. Since the 002 lattice fringes have also the advantage to resolve the problem of a precise
na
localization of the oriented areas [8], the above structural data are yielded directly by 002 lattice fringes.
Jo ur
It has to be mentioned that the data in this work were obtained by statistical data, the value can't accurately reveal the real microcrystalline size, however, they can illustrate the trend of the evolution of the microcrystalline size of pyrocarbon. The structural data are shown in Table. 1.
The OA values are derived from their SAED patterns. The OA value of the original HT is 47 ± 3°, while the OA values of the original MT is 70 ± 3°. The OA values of HT treated at 1600 °C, 2100 °C and 2450 °C are 38 ± 3°, 30 ± 3°, and 25 ± 3° respectively, while the OA values of MT treated at 1600 °C, 2100 °C and 2450 °C are 63 ± 3°, 53 ± 3°, and 50 ± 3°, respectively. It is definitely indicated that the orientation of carbon layers in HT and MT becomes preferable with the increase of treatment temperature. Undoubtedly, the OA values are more convincing to prove the evolution of the carbon layer at the
Journal Pre-proof
micro-level compared to Ae. Meanwhile, it can be found that the d002 of HT decreases from 0.3410 nm to 0.3361 nm, and the d002 of MT decreases from 0.3418 nm to 0.3372 nm. In addition, the value of L2 increases only a little, from 3.86 nm to 5.91 nm in HT and from 3.77 nm to 4.87 nm in MT. The value of L1 however, changes distinctly from 3.40 nm to 10.70 nm in HT and from 3.24 nm to 8.87 nm in MT as shown in Table 1, which means that the increasing tendency of the microcrystalline size of HT is greater
of
than that of MT. In other words, based on the transformation of microcrystalline size, it is distinctly
ro
demonstrated that the size of the microcrystal structure of HT is all larger than that of MT with different
-p
temperature treatment.
XRD and Raman analysis are employed to verify the transformation. XRD patterns and analysis
re
results of HT and MT after different temperature treatment are shown in Fig. 9. It is apparent that all
lP
specimens exhibit a sharp (002) peak, the FWHM of HT is narrower than that of MT, and the 2θ of (002) in two textures shifts to right slightly as the temperature increases, indicating the increase of L c and g of
na
two textures. In addition, the increasing tendency of HT is more distinct than that of MT. Furthermore, by fitting the scatterplot of the angle shift (△θ) from XRD versus variation of interplanar spacing (△d) from
the formula:
Jo ur
HR-TEM (Table 1), as shown in Fig. 10, the variation of △d increases in proportion to △θ and satisfies
𝑦 = 0.61 + 9.34𝑥
(1)
where y is the variation of interplanar spacing value along the fitted curve and x is angle shift. It demonstrates that angle shift reflects directly the crystal size of microstructure. That is to say, the angle shift indirectly indicates the effect of temperature on the evolution of microstructure in C/C composite, which is consistent with the conclusion of HR-TEM. The Raman spectra and analysis results are presented in Fig. 11. There are two obvious vibration peaks in 800-2000 cm-1: the D peak at 1350 cm-1 reflects the defects within the carbon lattice; the G peak at 1580 cm-1 is used to demonstrate the degree of structural order [35-38]. It is not difficult to find that,
Journal Pre-proof
both in HT and MT C/C composites, the FWHM of D peak and G peak becomes narrower as the treatment temperature increases (Fig. 11(a-h)), indicating the increase of crystallinity of two textures. The Raman spectra are fitted with five Lorenz curves located about 1200 cm-1, 1350 cm-1, 1500 cm-1, 1580 cm-1 and 1620 cm-1, respectively [23]. The coefficient R (ID/IG) is obtained by the ratio of the integral intensity of D peak and G peak to characterize graphitization degree [39], and the La value has an inverse proportion relationship with R value, i.e., La=4.4R-1. The results of R (ID/IG) and La are shown in Fig. 11 (i) and (j) respectively. It can be seen that the decrease of R and the increase of L a of HT after heat treatment
of
are more obvious than those of MT, indicating higher graphitization of HT compared to MT at every temperature. The above phenomenon presents that the type of carbon bonding in HT and MT has
ro
transformed from sp3 to sp2 as the stacking method in carbon atoms changes. Meanwhile, from the
than that of MT after different heat treatment.
-p
changing tendency of the R value, it is further explained that the structural order degree of HT is greater
re
4.3 The influence of microstructural evolution on micro-hardness
lP
Generally, high temperature provides driving energy for rearrangement of carbon atoms, leading to the tortuous carbon eventually transformed into the graphite. Oberlin et al. [40] introduced that the
na
turbostratic structure started to disappear at a temperature range of 1500 °C-1700 °C and formed distorted layers. Hence, only a few of carbon atoms in the turbulence structure were rearranged at 1600 °C. The
Jo ur
rearrangement of carbon atoms further enhanced above 2100 °C, and more carbon atoms were rearranged regularly up to 2450 °C, as shown in Fig. 12(a). Surely, the process of atomic rearrangement corresponds to the transformation of microstructure in pyrolytic carbon, and the pyrolytic carbon in the as-deposited state contains a large amount of C-C bonds (σ-bonds) with a bond angle of 109.5°. It is quite easy to form defects, thus, the carbon layers are disorderly arranged in the as-deposited state. And the hydrogen release in pyrolytic carbon occurs at high temperature [27]. The C-C single bonds decrease, while C=C double bonds increase with the increasing temperature shown in Fig. 12(b). The shape of C=C double bonds is similar to a benzene ring whose carbon atoms are in the same plane, and this bond is also called conjugate big π-bond with a stable structure, which is beneficial to the orientation of carbon atoms. Moreover, because the HT includes more C=C bonds than MT, the carbon layers of HT are more ordered than that in MT in the as- deposited state. Therefore, the orientation of the carbon layer in HT is greater than MT at all heat-treatment temperature. It is consistent to the observation of HRTEM images shown in Fig. 6.
Journal Pre-proof
Since the microstructures of PyC is the main factor influencing the mechanical properties of C/C composites, the mechanical properties vary accompanied by the microstructural evolution. There is no exception of micro-hardness and modulus for HT and MT. According to the report of Lieberman et al [41], the HT matrix was relatively soft compared with MT matrix attributed to the difference in internal structure. As shown in Fig. 13, since the carbon layers in the MT are curled and stacked, the carbon layers in MT are tightly entangled with each other resulting in strong resistance when the indenter is pressed into the matrix. The carbon layers in HT are better aligned, which is good for slipping. Therefore, the carbon
of
layers are compressed directly when the force is applied. Hence, the resistance to external forces of MT is greater than that of HT, and the micro-hardness of the MT is larger than HT. In addition, the elastic
ro
modulus of MT is also higher than that of HT because of the relatively small deformation. Based on the
re
and HT specimens after heat treatment.
-p
microstructural evolution of HT and MT mentioned above, this conclusion is also applicable to the MT
5 Conclusions
lP
1. After treatment at different temperatures, the indentation depth of HT is always deeper than that of
na
MT, and the corresponding micro-hardness and elastic modulus of MT are also greater than that of HT. 2. The delamination of MT and HT was different after heat treatment. HT matrix was directly
Jo ur
delaminated into many nano-layers accompanied by multiple microcracks. However, MT matrix was delaminated into micro-layers in which there were lots of microcracks and nano-layers. 2. The interface width of HT is always narrower than that of MT, and there are numerous nanopores in MT even after heat treatment at 2450 °C. 4. The bridging phenomenon is markedly observed in HT and the d 002 of the bridging lamella with a preferable orientation is about 0.3359 nm. 5. A fitted formula is proposed to describe the relationship between the variation of interplanar spacing from HR-TEM and the angle shift from XRD, indicating that interplanar spacing is in proportion to angle shift. Declaration of Interest
Journal Pre-proof
There are no interests to declare. Acknowledgments We would like to thank the Analytical & Testing Center of Northwestern Polytechnical University for providing TEM technical support. The financial supported by the National Natural Science Foundation of China under Grant Nos.51821091, 51432008. References
of
[1] E. Fitzer. The future of carbon-carbon composites. Carbon, 1987, 25(2): 163-190.
ro
[2] J.E. Sheehan, K.W. Buesking, B.J. Sullivan. Carbon-carbon composites. Annual Review of Materials Science, 1994, 24(1): 19-44.
-p
[3] H.J. Li, A.J. Li, R.C. Bai, K.Z. Li. Numerical simulation of chemical vapor infiltration of propylene
re
into C/C composites with reduced multi-step kinetic models. Carbon, 2005, 43(14): 2937-2950. [4] Q.L. Shen, Q. Song, H.J. Li, C.X. Xiao, T.Y. Wang, H.J. Lin, W. Li. Fatigue strengthening of
lP
carbon/carbon composites modified with carbon nanotubes and silicon carbide nanowires. International Journal of Fatigue, 2019, 124: 411-421.
na
[5] Q. Song, K.Z. Li, H.J. Li, S.Y. Zhang, L.H. Qi, Q.G. Fu. A novel method to fabricate isotropic pyrocarbon by densifying a multi-walled carbon nanotube preform by fixed-bed chemical vapor
Jo ur
deposition. Carbon, 2013, 59: 547-550.
[6] R.J. Gray, J.V. Cathcart. Polarized light microscopy of pyrolytic carbon deposits. Journal of Nuclear Materials, 1966, 19(1): 81-89.
[7] X. Bourrat, B. Trouvat, G. Limousin, G. Vignoles, F. Doux. Pyrocarbon anisotropy as measured by electron diffraction and polarized light. Journal of Materials Research, 2000, 15(1): 92-101. [8] B. Reznik, K.J. Hüttinger. On the terminology for pyrolytic carbon. Carbon, 2002, 40(4): 621-624. [9] H.C. Cao, E. Bischoff, O. Sbaizero, M. Rühle, A.G. Evans, D.B. Marshall, J.J. Brennan. Effect of interfaces on the properties of fiber‐ reinforced ceramics. Journal of the American Ceramic Society, 1990, 73(6): 1691-1699. [10] T.P. Weihs, O. Sbaizero, E.Y. Luh, W.D. Nix. Correlating the mechanical properties of a continuous fiber‐ reinforced ceramic‐ matrix composite to the sliding resistance of the fibers. Journal of the
Journal Pre-proof
American Ceramic Society, 1991, 74(3): 535-540. [11] S.J. Park, M.H. Kim. Effect of acidic anode treatment on carbon fibers for increasing fiber-matrix adhesion and its relationship to interlaminar shear strength of composites. Journal of Materials Science, 2000, 35(8): 1901-1905. [12] F.H. Zhang, R.G. Wang, X.D. He, C. Wang, L.N. Ren. Interfacial shearing strength and reinforcing mechanisms of an epoxy composite reinforced using a carbon nanotube/carbon fiber hybrid. Journal of Materials Science, 2009, 44(13): 3574-3577.
of
[13] Q. Song, K.Z. Li, L.H. Qi, H.J. Li, J.H. Lu, L.L. Zhang, Q.G. Fu. The reinforcement and toughening
ro
of pyrocarbon-based carbon/carbon composite by controlling carbon nanotube growth position in carbon felt. Materials Science & Engineering A Structural Materials Properties Microstructure &
-p
Processing, 2013, 564: 71-75.
re
[14] X.J. Chao, L.H. Qi, W.L. Tian, X.H. Hou, W.J. Ma, H.J. Li. Numerical evaluation of the influence of porosity on bending properties of 2D carbon/carbon composites. Composites Part B: Engineering,
lP
2018, 136: 72-80.
1998.
na
[15] E. Fitzer, L.M. Manocha, Carbon reinforcements and carbon/carbon composites. Springer, Berlin,
[16] W.G. Zhang, Z.J. Hu, K.J. Hüttinger. Chemical vapor infiltration of carbon fiber felt: optimization of
Jo ur
densification and carbon microstructure. Carbon, 2002, 40(14): 2529-2545. [17] H.O. Pierson, D.A. Northrop. Carbon-felt, carbon-matrix composites: dependence of thermal and mechanical properties on fiber precursor and matrix structure. Journal of Composite Materials 1974, 9: 118-137.
[18] R. Piat, Y. Lapusta, T. Böhlke, M. Guellali, B. Reznik, D. Gerthsen, T.F. Chen, R. Oberacker, M.J. Hoffmann. Microstructure-induced thermal stresses in pyrolytic carbon matrices at temperatures up to 2900° C. Journal of the European Ceramic Society, 2007, 27(16): 4813-4820. [19] S. Ozcan, J. Tezcan, B. Gurung, P. Filip. The effect of heat treatment temperature on the interfacial shear strength of C/C composites. Journal of Materials Science, 2010, 46(1): 38-46. [20] K. Fujita, H. Sakai, N. Iwashita, Y. Sawada. Influence of heat treatment temperature on interfacial shear strength of C/C. Composites Part A: Applied Science and Manufacturing, 1999, 30(4):
Journal Pre-proof
497-501. [21] L.H. Xia, B.Y. Huang, F.Q. Zhang, Z.M. Liu, T.F. Chen. Effect of heat treatment on cracking and strength of carbon/carbon composites with smooth laminar pyrocarbon matrix. Materials & Design, 2016, 107: 33-40. [22] W. Li, H.J. Li, J. Wang, S.Y. Zhang, X. Yang, J.F. Wei. Preparation and mechanical properties of carbon/carbon composites with high textured pyrolytic carbon matrix. Transactions of Nonferrous Metals Society of China, 2013, 23(7): 2129-2134.
of
[23] G.H. Zhou, Y.Q. Liu, L.L. He, Q.H. Guo, H.Q. Ye. Microstructure difference between core and skin
ro
of T700 carbon fibers in heat-treated carbon/carbon composites. Carbon, 2011, 49(9): 2883-2892. [24] A. Tressaud, M. Chambon, V. Gupta, S. Flandrois, O.P. Bahl. Fluorine-intercalated carbon fibers III.
-p
A transmission electron microscopy study. Carbon, 1995, 33(9): 1339-1345.
re
[25] W. Yang, R.Y. Luo, Z.H. Hou, Y. Zhang, H.D. Shang, M.Y. Hao. Influence of the microstructure of the carbon matrices on the internal friction behavior of carbon/carbon composites. New Carbon
lP
Materials, 2016, 31(2): 159-166.
[26] M.L. Li, L.H. Qi, H.J. Li, G.Z. Xu. Measurement of the extinction angle about laminar pyrocarbons
na
by image analysis in reflection polarized light. Materials Science and Engineering: A, 2007, 448(1-2): 80-87.
Jo ur
[27] B. Reznik, K. Norinaga, D. Gerthsen, O. Deutschmann. The effect of cooling rate on hydrogen release from a pyrolytic carbon coating and its resulting morphology. Carbon, 2006, 44(7): 1330-1334.
[28] B. Reznik, M. Fotouhi. Electron microscopy and electron-energy-loss spectroscopy study of crack bridging in carbon–carbon composites. Composites Science & Technology, 2008, 68(5): 1131-1135. [29] J.J. Ren, K.Z. Li, S.Y. Zhang, X.Y. Yao, H.J. Li. Preparation of high texture three-dimensional braided carbon/carbon composites by pyrolysis of ethanol and methane. Ceramics International, 2016, 42(2): 2887-2891. [30] F. Cancino-Trejo, M. Sáenz Padilla, E. López-Honorato, U. Carvajal-Nunez, J. Boshoven, J. Somers. The effect of heat treatment on the microstructure and diffusion of silver in pyrolytic carbon coatings. Carbon, 2016, 109: 542-551.
Journal Pre-proof
[31] A. Pfrang, B. Reznik, D. Gerthsen, T. Schimmel. Comparative study of differently textured pyrolytic carbon layers by atomic force, transmission electron and polarized light microscopy. Carbon, 2003, 41(1): 181-185. [32] B. Reznik, D. Gerthsen, K.J. Hüttinger. Micro- and nanostructure of the carbon matrix of infiltrated carbon fiber felts. Carbon, 2001, 39(2): 215-229. [33] F.Q. Zhang, B.Y. Huang, Q.Z. Huang, X. Xiong, T.F. Cheng. Effects of the interface on the graphitization of a carbon fiber/pyrolytic carbon composite. Carbon, 2003, 41(3): 610-612.
of
[34] A. Oberlin. Pyrocarbons. Carbon, 2002, 40(1): 7-24.
graphene. Applied Physics Letters, 2008, 92(4): 666.
ro
[35] Y.Y. Wang, Z.H. Ni, Z.X. Shen, H.M. Wang, Y.H. Wu. Interference enhancement of Raman signal of
-p
[36] D. Chotikapanich, W.E. Griffiths. Estimating Lorenz curves using a Dirichlet distribution. Journal of
re
Business & Economic Statistics, 2002, 20(2): 290-295.
[37] Q. Song, H.B. Yan, K.D. Liu, K.Y. Xie, W. Li, W.H. Gai, G.H. Chen, H.J. Li, C. Shen, Q.G. Fu.
lP
Vertically grown edge‐ rich graphene nanosheets for spatial control of Li nucleation. Advanced Energy Materials, 2018, 8(22): 1800564.
Lightweight
and
na
[38] X.M. Yin, H.J. Li, L.Y. Han, J.C. Meng, J.H. Lu, L.L. Zhang, W. Li, Q.G. Fu, K.Z. Li, Q. Song. flexible
graphene
shielding.
Jo ur
electromagnetic-interference
3D
microtubes Chemical
membrane Engineering
for
high-efficiency
Journal,
2020:
10.1016/j.cej.2020.124025.
[39] C. Casiraghi, S. Pisana, K.S. Novoselov, A.K. Geim, A.C. Ferrari. Raman fingerprint of charged impurities in graphene. Applied Physics Letters, 2007, 91(23): 183. [40] A. Oberlin. carbonization and graphitization. Carbon, 1984, 22(6): 521-541. [41] M.L. Lieberman, H.O. Pierson. Effect of gas phase conditions on resultant matrix pyrocarbons in carbon/carbon composites. Carbon, 1974, 12(3): 233-241.
Journal Pre-proof
Declaration of interests The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Jo ur
na
lP
re
-p
ro
of
☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:
Journal Pre-proof
Credit Author Statement
Jo ur
na
lP
re
-p
ro
of
Tiyuan Wang: Methodology, software, investigation, formal writing-original draft preparation; Hejun Li: Conceptualization, resources and funding acquisition; Shouyang Zhang: writing-review and editing, Conceptualization; Kun Li: Visualization, supervision and project administration; Wei Li: Data curation, Validation.
analysis
and
Journal Pre-proof
of
Fig. 1 Polarized light micrographs of HT(a-d) and MT(e-h) with different treatment: (a, e)as-deposited
Jo ur
na
lP
re
-p
ro
state; (b, f)1600°C; (c, g)2100°C; (d, h)2450°C
of
Journal Pre-proof
Jo ur
na
lP
re
-p
ro
Fig. 2 The fitted extinction curves and the corresponding extinction angle of HT(a) and MT(b).
Jo ur
na
lP
re
-p
ro
of
Journal Pre-proof
Fig. 3 SEM images of HT(a, c, e, g) and MT(b, d, f, h) with different treatment: (a, b)as-deposited state; (c, d)1600°C; (e, f)2100°C; (g, h)2450°C.
Jo ur
na
lP
re
-p
ro
of
Journal Pre-proof
Journal Pre-proof
Fig. 4 TEM images of HT(a, c, e, g) and MT(b, d, f, h) with different treatment: (a, b)as-deposited state;
-p
ro
of
(c, d)1600°C; (e, f)2100°C; (g, h)2450°C.
Figure.5 The surface morphology of carbon fiber (a) and the interface width of HT and MT under four
Jo ur
na
lP
re
temperatures (b).
Jo ur
na
lP
re
-p
ro
of
Journal Pre-proof
Jo ur
na
lP
re
-p
ro
of
Journal Pre-proof
Fig. 6 HR-TEM micrographs of HT(a, c, e, g) and MT(b, d, f, h) with different treatment: (a, b)as-deposited state; (c, d)1600°C; (e, f)2100°C; (g, h)2450°C.
Journal Pre-proof
Fig. 7 The bridging image in HT:(a)low-magnification image;(b)overview of microcrack; (c)HR-TEM
Jo ur
na
lP
re
-p
ro
of
image of (b).
lP
re
-p
ro
of
Journal Pre-proof
na
Fig. 8 Micro-indentation curves and results of two textures with different treatment: (a)micro-indentation
Jo ur
curves of HT; (b)micro-indentation curves of MT; (c)micro-hardness results; (d)elastic modulus results.
lP
re
-p
ro
of
Journal Pre-proof
Fig. 9 XRD patterns and analysis results of HT and MT with different treatment: (a) XRD patterns of HT;
Jo ur
na
(b)XRD patterns of MT; (c) Lc results of HT and MT; (d) graphitization degree of HT and MT.
re
-p
ro
of
Journal Pre-proof
Jo ur
na
lP
Fig. 10 Plot of angle shift (△θ) from XRD versus variation of interplanar spacing (△d) from HR-TEM
Jo ur
na
lP
re
-p
ro
of
Journal Pre-proof
Fig. 11 Raman spectra of HT(a, c, e, g) and MT(b, d, f, h) with different treatment: (a, b)as-deposited state;
Journal Pre-proof
lP
re
-p
ro
of
(c, d)1600°C; (e, f)2100°C; (g, h)2450°C; and (i)R(ID/IG) results; (j) La results.
Fig. 12 Sketches of the different microstructures of MT and HT (a) and the transformation of bonds
Jo ur
na
before and after HTT(b).
lP
re
-p
ro
of
Journal Pre-proof
Jo ur
na
Fig. 13 Schematic diagram of carbon layers change before and after indentation.
Journal Pre-proof
Table.1 Statistical structural data obtained from HR-TEM images. L1/(nm)
L2/(nm)
OA/°
HT
0.3410±0.001
3.40±0.04
3.86±0.05
47±3°
HT-1600°C
0.3396±0.002
4.96±0.05
4.14±0.03
38±3°
HT-2100°C
0.3365±0.001
6.98±0.003
4.49±0.04
30±3°
HT-2450°C
0.3361±0.001
10.70±0.05
5.91±0.04
25±3°
MT
0.3418±0.002
3.24±0.04
3.77±0.05
70±3°
MT-1600°C
0.3404±0.001
4.78±0.05
3.88±0.04
64±3°
MT-2100°C
0.3387±0.001
5.94±0.04
4.11±0.03
53±3°
MT-2450°C
0.3372±0.002
8.87±0.05
4.87±0.05
50±3°
Jo ur
na
lP
re
-p
ro
of
d002/(nm)
Journal Pre-proof
of
Graphical abstract
Jo ur
na
lP
re
-p
ro
Graphical abstract
Journal Pre-proof
Highlights 1.In-depth observation of microstructural evolution of different pyrolytic carbon
2.The microstructural distinctions between different pyrocarbon were compared at multiple scales
3.The influence of microstructural evolution on the micromechanics of composites was
Jo ur
na
lP
re
-p
ro
of
analyzed