Polymer precursor synthesis of TaC–SiC ultrahigh temperature ceramic nanocomposites

Abstract

TaC–SiC ultrahigh temperature ceramic nanocomposites were prepared by cross-linking and subsequent pyrolysis of a novel soluble blend precursor in argon atmosphere. The precursor was synthesized by blending polycarbosilane (PCS) with polytantaloxane (PT). The geometry of metal–ligand interaction of polymer PT was determined by the X-ray absorption fine structure (XAFS). The prepared ceramic nanomaterials were investigated with respect to their chemical and phase composition, by means of FT-IR, TG, X-ray diffraction (XRD), Micro-Raman spectroscopy, scanning electron microscopy (SEM) and high-resolution transmission electron microscopy (HR-TEM). Annealing experiments of the TaC–SiC samples at temperatures in the range of 1000–1800 °C showed conversion into nanostructured ultrahigh temperature ceramic composites with a trace amount of free carbon. The average grain sizes of the precursor-derived TaC and SiC ceramics were both less than 50 nm. Ta, Si and C elements were homogeneously distributed in the sample at the submicron scale.

---

Introduction

Ultrahigh temperature ceramics (UHTCs) have found promising applications in sharp wing leading edges (WLEs) and nose tips for thermal protection of hypersonic vehicles under re-entry conditions,¹ due to their exceptional combination of high melting point, high thermal conductivity and excellent mechanical properties.^2,3 TaC was one of the most outstanding ultrahigh temperature ceramics, with high hardness, high elastic modulus and chemical stability even in harsh environments.⁴ However, elevated temperatures induce grain growth, which impairs densification of the ceramic monoliths and thus the mechanical properties of the ceramic parts.^3,5,6 Meanwhile, similar to other UHTCs systems,^7–9 TaC suffers from rather poor oxidation resistance at high temperatures.¹⁰

One effective approach to address the problem mentioned-above is to introduce a silica former component (e.g. SiC, Si₃N₄, MoSi₂ etc.) into UHTCs.¹¹ In this case, the grain size can be reduced to nano scale owing to the mutual crystallization inhibitions among multiphase and the oxidation resistance can be also improved concerning the formation of the dense silica-based glass layer at the surface, which acts as a protective layer with self-healing effect upon exposure to aggressive environments.^12–14

Recently, polymer precursor synthesis methods have raised increasing interests to generate as-mentioned ceramic nanocomposites,^15–18 for the precursors can be tailored at the molecular level in order to generate ceramic nanocomposites with homogeneous element distribution and decreased crystal size and consequently with improved properties.¹⁹ Li et al. prepared nanocrystalline ZrB₂–ZrC–SiC ceramics through a one-pot reaction of polyzirconoxanesal with boric acid and poly(methylsilylene)ethynylene. The average size of the ZrB₂, ZrC and SiC grains from the precursor heat-treated at 1400 °C was approximately 100 nm and the presence of SiC obviously restrained the oxidation of ZrC and ZrB₂ at 1000 °C.²⁰ Jia Yuan et al. reported precursor synthesis of hafnium-containing ultrahigh temperature ceramic nanocomposites (UHTC-NCs). Amorphous SiHfBCN ceramics were prepared from a commercial polysilazane (HTT 1800, AZ-EM), which was modified upon reactions with Hf(NEt₂)₄ and BH₃·SMe₂, and subsequently cross-linked and pyrolyzed.²¹ Qingbo Wen et al. synthesized SiC–HfC_xN_1−x-based ultrahigh temperature ceramic nanocomposite from a novel precursor prepared by the reaction of an allyl hydrido polycarbosilane (SMP10) and tetrakis(dimethylamido)hafnium(IV) (TDMAH). The average grain size of both HfC_0.83N_0.17 and SiC phases was found to be less than 100 nm.¹⁹ Yet as far as we know, there is no report on the precursor synthesis of TaC–SiC ultrahigh temperature ceramic nanocomposites so far.

So in the present work, TaC–SiC ceramic nanocomposites were successfully prepared from a soluble precursor synthesized through blending polycarbosilane (PCS) with polytantaloxane (PT). The polymer-to-ceramic transformation and the nano/microstructural evolution of the final TaC–SiC ceramics were also assessed. The presented results emphasize a convenient preparative approach to nanostructured ultrahigh-temperature stable materials (UHTC-NCs) starting from a greatly flexible precursor.

Experimental

Synthesis

All chemicals were used as received. Tantalum alkoxide was prepared by the reaction between TaCl₅ and n-butnol (molar ratio of 1/5) with triethylamine as co-precipitating agent. After filtration, the filtrate was mixed in acetylacetone (Hacac) at 60 °C with Ta/Hacac molar ratio of 1/2.2. After refluxing the mixture for 1 h, 0.58 mol deionized water was added drop wise at 60 °C, followed by refluxing the mixture for another 1 h and adding 300 g xylene in order to better solve PCS. Then after distilling the mixture until the temperature reached 137 °C, PT in form of a dark brown viscous solution was obtained. A commercially available PCS was dissolved in xylene with mass ratio of 1/1. The precursor for TaC–SiC named PTS was synthesized by blending this solution with as-prepared PT solution. The PTS precursors with Ta/Si mass ratio of 2/1, 1/1 and 1/2 were marked as PTS-21, PTS-11 and PTS-12, respectively.

The obtained precursors were cured at 150 °C (heating rate of 5 °C min⁻¹, dwelling time of 2 h) and then at 260 °C (heating rate of 10 °C min⁻¹, dwelling time of 2 h) to produce brown red solids. The cured samples were heat-treated in argon atmosphere from room temperature to 1000 °C, 1200 °C, 1400 °C, 1500 °C, 1600 °C and 1800 °C respectively, with a heating rate of 3 °C min⁻¹ and dwelling time of 2 h. Finally, the samples were cooled to room temperature and TaC–SiC ceramic powders were obtained.

Characterization

XAFS measurements were performed at Institute of High Energy Physics, Chinese Academy of Sciences to characterize the geometry of metal–ligand interaction of polymer PT. FT-IR was measured to analyse the molecule structure of the precursor. Powder X-ray diffraction (XRD), Raman spectroscopy, thermal gravimetric analysis (TGA), scanning electron microscopy (SEM) and high resolution transmission electron microscopy (HR-TEM) were used to characterize the prepared ceramic materials. Phase identification was carried out by a Rigaku D/MAX-2400 XRD powder diffractometer (Japan) with Cu-Kα radiation. The diffraction data were collected within 2θ range of 3–80° in steps of 8° min⁻¹. Raman spectra were recorded from 800 cm⁻¹ to 3200 cm⁻¹ employing a micro-Raman spectrometer (Renishaw in Via plus, Britain) using a laser wavelength of 633 nm. The thermal behavior of the cured precursors was investigated by thermal gravimetric analysis (TGA) (Mettler Toledo TGA/SDTA851, German). The thermal analysis included heating under flowing argon atmosphere with a rate of 10 °C min⁻¹ ranging from room temperature to 1500 °C and a flow speed of 50 ml min⁻¹. Ta and Si elemental analysis was carried out by Variant inductively coupled plasma emission (ICP) spectrometry (Thermo IRIS intrepid II) and carbon content was measured on carbon/sulfur analyser (CS-206, Baoying Technology). Oxygen content was investigated by oxygen analyser. The microstructure evolution of the ceramic samples varying with temperature was observed by Hitachi S-4800 scanning electron microscopy (SEM, Hitachi S-4800, Japan) in secondary electron mode. The distribution of elements was observed by mapping on energy-dispersive X-ray microanalysis attachments of the Hitachi S-4300 scanning electron microscopy (Japan). Nanostructure of the ceramic samples was studied by high-resolution transmission electron microscopy (HR-TEM) on ground powder samples using a JEM-2100F (Japan) microscope at an acceleration voltage of 200 kV (wavelength 152.51 pm). In addition, the selected area electron diffraction (SAED) technique was employed in order to investigate the phase composition in nano scale.

Results and discussions

The structure of the precursor

Since Ta-based polymer shows unusual complexity to be characterized by common measurements for organic compounds such as NMR, MS and the absence of single crystal, XAFS was introduced into the analysis of structure of Ta-based polymer therefore. The oxidation state of tantalum was determined by comparison of the absorption edge position of the PT polymer and the reference oxide as shown in Fig. 1. The study indicated the +5 oxidation state of Ta ions in the analysed compound.

Fig. 1 Comparison of X-ray absorption near edge structure spectra for PT polymer and reference oxide.

EXAFS analysis shown in Fig. 2(a) revealed that the first coordination sphere around core tantalum atom consists of six oxygen atoms. [(OEt)₃Ta(μ,μc-OBzP)]₂ (ref. 22)was selected from the database as model. The EXAFS oscillations and the fit were presented in Fig. 2. Fitting parameters were listed in Table 1. P. N. Kapoor studied that the reactions of tantalum pentaethoxides with excess (5 moles) of acetylacetone at room temperature yielded a tetraethoxymonoacetylacetonate which was found to be monomeric having therefore a coordination number of six.²³ Funk also found that a coordination number of six is the most common and stable for tantalum.²⁴ In our case, tantalum atom in PT polymer also adopted the stable six-fold coordination structure, which suggested acetylacetone was in monomeric coordination. The proposed molecular structure of PT polymer was shown in Fig. 2(b).

Fig. 2 Fourier-transformed experimental extended X-ray absorption fine structure oscillations for PT polymer and fitted coordination spheres (a) and proposed structure of PT polymer (hydrogen atoms were omitted) (b).

Table 1 The distance from central cation to the next atom (R), number of atoms (N) and R-factor of the fit evaluated during EXAFS fitting procedure^a

Sample	Path	R (Å)	N	R-factor
a EXAFS, extended X-ray absorption fine structure.
PT polymer	Ta–O	1.89	6.24	0.0003

---

Conjugate structure could be formed by coordination of acetylacetone to Ta atom in the reaction of tantalum alkoxides with acetylacetone,²⁵ which was also confirmed by the red shift of C=O (1590 cm⁻¹) and C=C (1533 cm⁻¹) in FT-IR spectrums of PTS precursors in Fig. 3(a). The coordination effect contributed to the stability of the liquid precursor, resulting in a long shell life of the precursor. FT-IR also revealed the absorption bands that can be assigned to the presence of Si–H (2100 cm⁻¹) and Si–CH₃ bonds (1250 cm⁻¹) from PCS. Absorption bands which could be related to Ta–O were measured between 480 cm⁻¹ and 520 cm⁻¹. The absorption at 1700 cm⁻¹ suggested some trace amount of free Hacac existing in the blend, which can better promote the coordination equilibrium reaction to stabilize the precursor to some degree.

Fig. 3 FT-IR spectrums of PTS precursors (a) and conjugate structure (b).

Thermal behavior of PTS precursor

To investigate the polymer-to-ceramic conversions, TGA and DTG of cured PTS-21, PTS-11 and PTS-12 precursors were measured, as shown in Fig. 4. No obvious mass loss was detected below 200 °C, for most volatile mixed in the precursors had been removed during the curing process. The organic-to-inorganic transformation mainly happened in the range from 250 to 700 °C, as indicated by the sharp decrease of mass. The weight-loss step which began at 1000 °C for PTS-21, 1200 °C for PTS-11 and 1300 °C for PTS-12, was attributed to the carbothermal reduction. By comparisons of the three PTS samples, it's obviously that the starting temperature of carbothermal reduction was postponed with increasing silicon former component, implying the interactions between silicon constituent and tantalum constituent.

Fig. 4 TGA (a) and DTG (b) curves of cured PTS precursors.

XRD was employed to understand the crystallization behavior during pyrolytic conversion of PTS precursor. Fig. 5(a) showed a typical XRD pattern of samples pyrolyzed from PTS-11 precursor at various temperatures. 1000 °C heat-treated sample of PTS-11 precursor was nearly amorphous, while PT-derived TaC was already crystalized at 900 °C.²⁶ At 1200 °C, obvious β-Ta₂O₅ peaks (PDF #25-922) emerged as well as weak TaC peaks (PDF #65-282) appeared, implying the beginning of carbothermal reduction, which was consistent with the TGA results of the mass loss at above 1000 °C. With temperature rising, β-Ta₂O₅ peaks decreased and TaC peaks increased. This is because β-Ta₂O₅ was gradually consumed by reaction with the free carbon to form TaC in the process of carbothermal reduction. TaC became the predominant phase and β-SiC (PDF #49-1623) was firstly detected when the temperature was elevated to 1500 °C, indicating the oxides have been transformed into corresponding TaC via carbothermal reduction. Further increasing heat treatment temperature, the diffraction peaks of TaC and SiC became sharper and stronger, indicating higher crystalline degree. Overall, compared to the initial crystallization temperatures of PT-derived TaC ceramics (900 °C) and that of PCS-derived SiC ceramics (1200 °C), the initial crystallization temperatures of both TaC and SiC in the PTS-derived ceramics were postponed. That reflected the crystalline inhibition between the TaC phase and SiC phase thereby. Elemental analysis of ceramic powders derived from PTS-11 at 1600 °C: Ta 38.7 wt%, Si 39.4 wt%, C 20.5 wt%, O 0.4 wt%. Since free carbon is detrimental for ultrahigh temperature performances, the excess carbon content as low as 1 wt% in this case showed promising potentials for ultrahigh temperature applications.

Fig. 5 XRD pattern of samples pyrolyzed from PTS-11 precursors pyrolyzed at various temperatures (a) and samples from different PTS precursors pyrolyzed at 1600 °C (b).

To examine the interaction between silicon constituent and tantalum constituent more thoroughly, XRD patterns of samples from different PTS precursors pyrolyzed at 1600 °C were measured for comparison, shown in Fig. 5(b). TaC and SiC crystalline phases were all detected in the three samples. Among the three samples, however, with increasing Si content in PTS precursors, the precursor-derived TaC crystalline peaks became less sharp, implying crystallization of TaC was inhibited by SiC. Vice versa, with increasing Ta content in PTS precursors, SiC crystalline peaks also turned less sharp, indicating crystallization of SiC was similarly inhibited by TaC. Further estimation of average crystallite sizes from the XRD spectrum using Debye–Scherrer equation showed that the average crystallite size of TaC was 26–35 nm, while that of β-SiC was around 21–25 nm. The reduced crystallite size to nano scale was closely related to the inhibition of crystallization between the two crystalline phases and the homogeneous elements distribution from the intrinsic properties of polymer precursor, and thus TaC–SiC nanocomposites could be obtained.

Carbon phase

Micro Raman spectroscopy, as one of the most sensitive methods for characterization of the different modifications of carbon, was utilized to identify the evolution of free carbon in the ceramics. As shown in Fig. 6, there are two obvious peaks centered around 1323 and 1595 cm⁻¹ for samples obtained from PTS-11 at 1000 °C, 1200 °C and 1400 °C, corresponding to D and G peaks of free carbon. The peak at 1323 cm⁻¹ is attributed to disordered and amorphous carbon bonds that originate from the breathing mode of k-point phonons of asymmetry, while the peak at 1595 cm⁻¹ describes a stretching mode attributable to sp²-hybridized carbon atoms in the honeycomb network.²⁷

Fig. 6 Micro Raman spectra of samples prepared from PTS-11 precursor at various temperatures (a) and different PTS precursors at 1600 °C (b).

The degree of disorder was rationalized on the basis of the parameter L_D (interdefect distance), which has been defined, cf. I_D/I_G = C(λ)/L_D.^2,27 As a consequence, the average size of sp²-bonded graphitic domains is inversely proportional to the I_D/I_G ratio.²⁸ Then the I_D/I_G data were summarized in Table 2. With the annealing temperature rising, the ratio of I_D/I_G decreased, illustrating the L_D became larger. Thus, structural organization of the carbon phase increased as temperature went up. Note should also be put on the extremely weak D band and G band in 1600 °C and 1800 °C heat-treated samples, indicating the quite low free carbon content.

Table 2 Peak positions and integral area ratios (I_D/I_G) of the D and G modes for the segregated carbon phase present in TaC–SiC ceramics from PTS-11 annealed in argon atmosphere at different temperatures

Anealing temperature (°C)	ωD (cm^−1)	ωG (cm^−1)	ID/IG
1000	1323	1595	2.55
1200	1321	1586	1.91
1400	1323	1592	1.25

---

From the Raman patterns of samples derived from different PTS precursors at 1600 °C, we can see free carbon content increased with increasing PCS fraction in the precursor. PCS bears excess carbon for converting tantalum oxide into tantalum carbide by carbothermal reduction. With raising tantalum component, more carbon was consumed by this carbothermal reduction, which led to lower free carbon content in the final ceramic products.

Micro/nanostructure

The microstructure evolution of the annealed TaC–SiC ceramic samples was revealed by SEM analysis shown in Fig. 7. The sample was relatively compact below 1200 °C due to the significant volume shrinkage after the curing course, while at 1400 °C, the compact bulk started to crush due to carbothermal reduction. With temperature rising to 1500 and 1600 °C, the release of gas in the carbothermal reaction resulted in the formation of pores with different sizes, and the sample gradually converted into porous ceramics consequently. The ceramic particles showed near spherical morphology with particle size less than 200 nm. Raising temperature up to 1800 °C led to the grain coarsening, which rendered the ceramic particles aggregate into lump.

Fig. 7 SEM micrographs of samples prepared from PTS-11 precursors at (a) 1000 °C; (b) 1200 °C; (c) 1400 °C; (d) 1500 °C; (e) 1600 °C; (f) 1800 °C.

SEM images of ceramic samples pyrolyzed at 1600 °C derived from different PTS precursors were also shown in Fig. 8 to trace the microstructure changes with different silicon former constituents. PTS-21 derived ceramics were aggregated somehow, leading to the formation of larger particles resultantly, while aggregation and particle size of PTS-11 derived ceramics were both declined. The ceramics prepared from PTS-12 showed the least aggregation and the smallest particle size among the three samples. Therefore, the inhibition of grain growth and coarsening between TaC and SiC crystallines was confirmed again.

Fig. 8 SEM images of ceramic samples pyrolyzed at 1600 °C derived from (left) PTS-21; (middle) PTS-11; (right) PTS-12.

Ceramics derived from precursor solutions indicated homogeneous atomic distribution among different components.²⁹ Therefore, mapping of TaC–SiC ceramic sample from PTS-11 precursor annealed at 1600 °C was shown in Fig. 9 to investigate the distribution of elements. Ta, Si and C elements were all well distributed in submicron scale in the ceramic sample without large-scale phase separation detected. Hence the feature of elements homogeneity for precursor-derived ceramics was strongly verified with this respect.

Fig. 9 SEM image and elemental distribution of PTS-11 derived ceramic sample at 1600 °C (a) mapping zone, (b) Ta map, (c) Si map, and (d) C map.

HR-TEM is a powerful method to characterize the micro/nanostructure of ceramic nanocomposites. Thus, PTS-12 derived TaC–SiC ceramics synthesized at 1800 °C were studied by means of HR-TEM with respect to phase composition and micro/nanostructure (Fig. 10). Bright field lattice image (Fig. 10(a)) showed that dark TaC particles and light β-SiC particles were dispersed with near spherical morphology. The size of the nanocrystals was less than 50 nm. As found by selected area electron diffraction (SAED), the crystalline phases consist of β-SiC and TaC. Additionally, graphite-like carbon was also detected, which was consistent with the Raman results. Fig. 10(b) revealed the characteristic nanostructure of the 1800 °C ceramic sample, where crystallites of β-SiC and TaC dispersed uniformly with a little amount of amorphous carbon embedded in the crystal boundary area. Highly organic nanostructure revealed an enhanced crystallization. On the basis of HR-TEM results combined with XRD studies and SEM observations, the PTS precursors derived TaC–SiC were ceramic nanocomposites with a trace amount of free carbon, which was owing to the crystallization inhibition to a large degree.

Fig. 10 HR-TEM micrographs of ceramic sample prepared from PTS-12 precursor pyrolyzed at 1800 °C (inset: selected area electron diffraction).

Considering the obtained results, the TaC–SiC ceramic material prepared upon pyrolysis of the precursor was greatly versatile, with regard to its crystallization behavior, phase evolution and micro/nanostructure. Relatively low temperature annealing of the precursors at 1600 °C generated ceramic nanocomposites, which were expected to be promising candidates for applications at (ultra)high temperatures and under extreme environmental conditions.

Conclusion

Within the present study, a precursor preparative access toward TaC–SiC ultrahigh temperature ceramic nanocomposite has been discussed. The precursor was prepared by blending polycarbosilane with polytantaloxane in which Ta was coordinated in six-fold structure. Pyrolysis of the PTS-11 precursor at 1600 °C could produce TaC–SiC crystal phases with a trace amount of free carbon. Micro/nanostructure studies revealed the average grain size of the generated ceramic composites reached nano scale of 50 nm. The comprehensive characterizations highlighted the crystallization inhibition between the UHTC component and silica former component, which mainly contributed to the formation of nanostructured TaC–SiC ceramic composites in this work. The presented precursor synthesis approach is believed to be applicable for other phase compositions consisting of a UHTC phase such as group IV transition-metal borides, carbides and nitrides dispersed within a silica-former matrix such as SiC and Si₃N₄.

Acknowledgements

This project was supported by the National Basic Research Program of China (No. 2012CB933201).

References

1. E. Zapata-Solvas, D. D. Jayaseelan, H. T. Lin, P. Brown and W. E. Lee, J. Eur. Ceram. Soc., 2013, 33, 1373–1386.
2. W. G. Fahrenholtz, G. E. Hilmas, I. G. Talmy and J. A. Zaykoski, J. Am. Ceram. Soc., 2007, 90, 1347–1364.
3. L. Silvestroni and D. Sciti, J. Am. Ceram. Soc., 2011, 94, 1920–1930.
4. I. G. Talmy, J. A. Zaykoski and M. M. Opeka, J. Eur. Ceram. Soc., 2010, 30, 2253–2263.
5. X. Zhang, G. E. Hilmas and W. G. Fahrenholtz, Mater. Sci. Eng. A Struct. Mater., 2009, 501, 37–43.
6. S. A. Ghaffari, M. A. Faghihi-Sani, F. Golestani-Fard and H. Mandal, J. Eur. Ceram. Soc., 2013, 33, 1479–1484.
7. S. Shimada, M. Inagaki and K. Matsui, J. Am. Ceram. Soc., 1992, 75, 2671–2678.
8. P. Sarin, P. E. Driemeyer, R. P. Haggerty, D. K. Kim, J. L. Bell, Z. D. Apostolov and W. M. Kriven, J. Eur. Ceram. Soc., 2010, 30, 2375–2386.
9. W.-M. Guo and G.-J. Zhang, J. Eur. Ceram. Soc., 2010, 30, 2387–2395.
10. A. Nieto, A. Kumar, D. Lahiri, C. Zhang, S. Seal and A. Agarwal, Carbon, 2014, 67, 398–408.
11. J. W. Zimmermann, G. E. Hilmas and W. G. Fahrenholtz, Mater. Chem. Phys., 2008, 112, 140–145.
12. P. Hu, X.-H. Zhang, J.-C. Han, X.-G. Luo and S.-Y. Du, J. Am. Ceram. Soc., 2010, 93, 345–349.
13. H. Wu and W. Zhang, J. Eur. Ceram. Soc., 2010, 30, 1035–1042.
14. D. D. Jayaseelan, E. Zapata-Solvas, P. Brown and W. E. Lee, J. Am. Ceram. Soc., 2012, 95, 1247–1254.
15. E. Ionescu, H.-J. Kleebe and R. Riedel, Chem. Soc. Rev., 2012, 41, 5032–5052.
16. B. Papendorf, K. Nonnenmacher, E. Ionescu, H.-J. Kleebe and R. Riedel, Small, 2011, 7, 970–978.
17. H. Wu, C. Xie, W. Zhang and J. Zhang, Adv. Appl. Ceram., 2013, 112, 366–373.
18. C. Xie, M. Chen, X. Wei, M. Ge and W. Zhang, J. Am. Ceram. Soc., 2012, 95, 866–869.
19. Q. Wen, Y. Xu, B. Xu, C. Fasel, O. Guillon, G. Buntkowsky, Z. Yu, R. Riedel and E. Ionescu, Nanoscale, 2014, 6, 13678–13689.
20. Y. Li, W. Han, H. Li, J. Zhao and T. Zhao, Mater. Lett., 2012, 68, 101–103.
21. J. Yuan, S. Hapis, H. Breitzke, Y. Xu, C. Fasel, H.-J. Kleebe, G. Buntkowsky, R. Riedel and E. Ionescu, Inorg. Chem., 2014, 53, 10443–10455.
22. T. J. Boyle and L. A. M. Ottley, Inorg. Chim. Acta, 2010, 364, 69–80.
23. P. N. Kapoor and R. C. Mehrotra, J. Less-Common Met., 1965, 8, 339–346.
24. H. Funk, Berichte der deutschen chemischen Gesellschaft (A and B Series), 1934, 67, 1801–1804.
25. A. K. Narula, B. Singh, P. N. Kapoor and R. N. Kapoor, Synth. React. Inorg. Met.-Org. Chem., 1983, 13, 887–898.
26. Y. Lu, L. Ye, W. Han, Y. Sun, W. Qiu and T. Zhao, Ceram. Int., 2015, 41, 12475–12479.
27. E. H. Martins Ferreira, M. V. O. Moutinho, F. Stavale, M. M. Lucchese, R. B. Capaz, C. A. Achete and A. Jorio, Phys. Rev. B: Condens. Matter Mater. Phys., 2010, 82, 125429.
28. S. Eigler, C. Dotzer and A. Hirsch, Carbon, 2012, 50, 3666–3673.
29. T. Cai, W.-F. Qiu, D. Liu, W.-J. Han, L. Ye, A.-J. Zhao and T. Zhao, J. Am. Ceram. Soc., 2013, 96, 3023–3026.

---