Proceedings of the 12th Pacific Rim Conference on Ceramic and Glass Technology; Ceramic Transactions, Volume 264

Ceramic Transactions, Volume 264, Proceedings of the 12th Pacific Rim Conference on Ceramic and Glass Technology

Dileep Singh, Manabu Fukushima, Young-Wook Kim, Kiyoshi Shimamura, Nobuhito Imanaka, Tatsuki Ohji, Jake Amoroso, and Michael Lanagan; Editors

 

This proceedings contains a collection of 32 papers presented at the 12th Pacific Rim Conference on Ceramic and Glass Technology (PacRim12), May 21-26, 2017 in Waikoloa, Hawaii. PacRim is a bi-annual conference held in collaboration with the ceramic societies of the Pacific Rim countries - The American Ceramic Society, The Chinese Ceramic Society, The Korean Ceramic Society, and the Australian Ceramic Society. Topics included in this collection include multiscale modeling and simulation, processing and manufacturing, nanotechnology, multifunctional materials, ceramics for energy and the environment, biomedical materials, and more

• Language: English
• Publisher: Wiley
• Release date: May 2, 2018
• ISBN: 9781119494072

Book preview

Preface

The Pacific Rim Conference on Ceramic and Glass Technology is a bi-annual conference held in collaboration with the ceramic societies of the Pacific Rim countries—The American Ceramic Society, The Chinese Ceramic Society, The Korean Ceramic Society, and the Australian Ceramic Society. The 1st PACRIM conference was hosted by The American Ceramic Society (ACerS) at Maui, Hawaii, in 1993. Over the years, PACRIM conferences have earned a distinct reputation as a premier forum for presentations and discussions on the state-ofthe-art and emerging topics in ceramics and glass technologies.

The 12th Pacific Rim Conference on Ceramic and Glass Technology (PACRIM 12) was held at the Hilton Waikoloa Village, Waikoloa, Hawaii, May 21-26, 2017. PACRIM 12 had 34 Symposia and drew a record number of attendees—nearly 1,200—to give 1,400 presentations and 250 posters. Scientist, engineers, and students from 44 countries attended the conference, attesting to the truly international nature of the field of materials science, and ceramic and glass science in particular.

This Ceramic Transactions volume contains 32 peer reviewed papers from the following symposia held during PACRIM 12:

Symposium 2: Virtual Materials Design and Ceramic Genome

Symposium 3: Novel, Green, and Strategic Processing and Manufacturing Technologies

Symposium 4: Polymer Derived Ceramics (PDCS) and Composites

Symposium 6: Synthesis and Processing of Materials using Electric Currents and Pressures

Symposium 7: Porous Ceramics

Symposium 10: Multifunctional Nanomaterials and Their Heterostructures for Energy and Sensing Devices

Symposium 11: Engineering Ceramics: Processing and Characterizations

Symposium 12: Design, Development and Applications of Ceramic Matrix Composites

Symposium 13: Advanced Structural Ceramics for Extreme Environments

Symposium 16: Geopolymers: Low-Energy and Environmental-Friendly Ceramics

Symposium 17: Advanced Functional Ceramics and Critical Materials Perspective

Symposium 20: Crystalline Materials for Electrical, Optical and Medical Applications

Symposium 22: Direct Thermal to Electrical Energy Conversion Materials and Applications

Symposium 23: Materials for Solar Thermal Energy Conversion and Storage

Symposium 25: Ceramics for Next Generation Nuclear Energy

Symposium 27: Ceramics for Enabling Environmental Protection: Clean Air and Water

Symposium 29: Advances in Polar, Magnetic and Semiconductor Materials: Extending Temperature Limits

Symposium 32: Nanostructured Bioceramics and Ceramics for Biomedical Applications

Third International Richard M. Fulrath Symposium on Discontinuous Progress for Ceramic Innovations

Young Investigator Forum: Design and Application of Next Generation Multifunctional Materials

The editors wish to extend their gratitude and appreciation to all the symposium co-organizers for their help and support, to all the authors for their cooperation and contributions, to all the participants and session chairs for their time and efforts, and to all the reviewers for their valuable comments and suggestions. We hope that these proceedings will serve as a useful resource for materials scientists and engineers.

DILEEP SINGH

Conference Chair

PACRIM 12

Multiscale Modeling and Simulation

D⁰ FERROMAGNETISM OF SIC CERAMICS

Y. Huang, D. Jiang, Z. Huang

The State Key Laboratory of High Performance Ceramics and Superfine Structure, Shanghai Institute of Ceramics, Shanghai 200050, China

ABSTRACT

d⁰ ferromagnetism was presented in Al:SiC ceramics. Strong magnetic hysteresis loop and magnetic domain in Al:SiC ceramics was observed for the first time. Greenbody with oriented grains of magnetic Al:SiC was formed by slip casting in 6T magnetic field. After sintering, SiC ceramics were well-oriented, and magnetic domain was detected. First-principles calculations were used to calculate spin magnetic moment of SiC.

INTRODUCTION

It is well known that localized magnetic moments and the coupling between them are two indispensable factors to induce long-range spin ordering in solids, exhibiting ferromagnetism (FM), anti-ferromagnetism (AFM), ferrimagnetism. The local spins usually come from the elements containing the partially filled 3d or 4f subshells, where the electron configuration favors the high-spin states according to the Hund’s rule. The coupling strength between the spins depends on the exchange integral that is sensitive to the separation of the spins.

d⁰ ferromagnetism[1-4] is the property of materials, which have small ferromagnetic moments and Curie points well above room temperature despite the absence of atoms with partially filled d or f shells. Recently, there has been increasing evidence that traditional magnetic elements are not the sole source in inducing intrinsic magnetism; RT FMs were observed in highly oriented pyrolytic graphite (HOPG), in nonstoichiometric CaB6, in thin films of HfO2. SiC single crystals are otherwise nonmagnetic in their bulk states. But theoretical studies revealed that the local moment can form from defects and the extended tails of their wave functions mediate long-range magnetic coupling[5-8].

SiC is an important wide-gap semiconductor with more than 200 different polytypes in structures, which are sensitive to the fluctuation of experimental conditions. 6H-SiC has a hexagonal structure (P63mc) with lattice constants a= 0.3081 and c=1.5092 nm and a band gap of 3.0 eV at room temperature. Suzuki et al [9-10] utilized the anti-ferromagnetism of ceramics like SiC, Al2O3, Si3N4, et al to orient the grains of ceramics by slip casting in strong magnetic fields. Mao [11] used strong magnetic fields to make transparent Al2O3 ceramics due to its orientation at c axis. But all these papers [12-15] focused on anti-ferromagnetism of ceramics, and the values of the magnetic fields needed are always larger than 10T.

In this paper, we provide convincing experimental evidence that d⁰ ferromagnetism is presented in Al:SiC ceramics besides its single crystals. Strong magnetic hysteresis loop and magnetic domain in Al:SiC ceramics was observed for the first time. Greenbody with oriented grains of magnetic Al:SiC was formed by slip casting in 6T magnetic field. First-principles calculations were used to calculate spin magnetic moment of SiC.

EXPERIMENTAL

High-purity SiC powder and aluminum powder were used as precursors. Different contents of Al were added to obtain different level of doping in the samples. The doped samples were identified by the atomic percent of Al. Each mixed powder was loaded into a SiC crucible. The crucible was then transferred into a graphite furnace. The furnace was evacuated to low pressure (30Pa) before purged and filled with high purity argon. It was then heated to 2000oC and held for 2h. The as-synthesized powders were used to form greenbody by slip casting in 6T magnetic field. With the addition of B4C and C, the samples were sintered at 2180 oC for 1h with flowing argon.

The sample was put in a capsule for magnetic measurement. We characterized the sample structure by X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS). Magnetic domain of SiC ceramics were obtained by magnetic force microscopy (MFM). First principle calculations were performed by using the generalized gradient approximation in the form of the Perdew-Burke-Ernzerhof function, which was implemented in the Vienna ab initio simulation package.

RESULTS AND DISCUSSION

Graph shows range from gamma to H versus energy in eV from minus 16 to 8 where set of irregular curves are drawn with markings for Eg equals 2.2 eV and EF.

Fig. 1 Band structure calculation of 6H-SiC

Fig. 1 shows the energy band structure calculation of undoped 6H-SiC from first principles along the first Brillouin zone of high symmetry point G → A → H → K → G → M → L → H direction, the G point is the center of Brillouin zone. The valence electrons involved are C:2s2²p², Si:3s²3p², all calculation are performed in the reciprocal space. The top of the valence band is located in the center of the Brillouin Zone G, with the double degeneracy, following the energy of singlet. The bottom of the conduction band is at the K point. The calculated band gap is 2.2eV.

Graph shows energy in eV from minus 16 to 16 versus DOS from minus 1.6 to 1.6 where set of irregular curves are drawn with plots for total, C-s, C-p, Si-s, and Si-p.

Fig. 2 The electron spin density of the states of 6H-SiC

Fig. 2 is the density of electronic spin states in the 6H-SiC lattice, the total density of states of SiC and the partial atomic density of states were calculated. Calculation shows that: up and down spin density distribution is symmetry. The system shows no magnetism. Combined with Fig.1, valence band of 6H-SiC is mainly derived from 2p shell of C, and partially 3p shell of Si; the bottom of the conduction band is mainly composed of 3p shell of Si.

XRD results (Fig.3) showed that the phases of Al:SiC powder were 6H-SiC and 33R-SiC. Al4C3 peak was also detected at 78.3o. All 6H-SiC peaks were right shifted about 0.03 o, which means its cell was reduced. This was expected to be caused by Al replacing Si in the SiC cell.

Graph shows 2 theta/degree from 10 to 80 versus intensity from 0 to 2500 with plots for 6H-SiC, 33R-SiC, and Al4C3.

Fig. 3 XRD pattern of Al:SiC powder after calcination at 2000oC

According to the XRD results in Fig.3, and to understand the origin of the observed magnetism, a supercell consisting of 3 × 3 × 1 unit cells of 6H-SiC containing one AlSi-VSi, corresponding to a defect concentration of 0.93at%, was built for calculations (Fig.4). The results showed that the combination of Al and Vacancy lead to a local magnetic moment of 1.0 μ B. The up spin density was 3.95 electron/eV, while the down spin density was 3.15 electron/eV, which contributed to the net magnetic moment at fermi level.

Image shows set of atoms placed in certain pattern, and graph shows energy in eV from 0.0 to 1.5 versus density of states in electrons/eV from minus 4 to 4 with plots for alpha and beta.

Fig. 4 Sketch of defects in the supercell of 3 × 3 × 1 Al:SiC and its electron spin density of the states, in which one Si (yellow ball) atom is replaced with one Al (red ball) atom, one Si atom is replaced with Vacancy, and all C (gray ball) atoms remain in the same in defects.

Graph shows H(Oe) from minus 6000 to 6000 versus magnetization in emu/g from minus 0.012 to 0.012 where set of curves are drawn with plots for 1.0 percent Al:SiC, 0.5 percent Al:SiC, and commercial.

Fig. 5 The hysteresis loops for different Al doping SiC ceramics

In order to characterize the magnetism of the prepared powders, it is necessary to exclude the influence of d electronic elements first. Analysis by X ray fluorescence spectrum, there were not d shell elements like Fe, Co, Ni and f shell elements like rare earth in the prepared SiC powder and raw materials. Fig. 5 shows the magnetization curves for different Al content in SiC powder. Test temperature was 5K. It can be found that the ferromagnetic strengthen with the increase of Al content. When the Al content reaches 1%, the hysteresis phenomenon is obvious, both the area of hysteresis loop and the coercive force are impressive.

It is necessary to note that pure commercial SiC powder used in the experiment are impure. It can be found in Fig.5, the commercial SiC powder is also ferromagnetism in 5K (or paramagnetic), while defect free SiC should be anti-magnetism. From the magnetization curve, it may also generate d⁰ ferromagnetic phenomena due to the defect induced in the production process of commercial powder. For d⁰ ferromagnetism is very sensitive to defects. It is possible that d⁰ ferromagnetism had already been utilized in the reports of Suzuki et al. Unlike intrinsic diamagnetism, d⁰ ferromagnetism is controllable. Most of the material can be oriented at 10T under strong magnetic field, so magnetic type cannot be identified.

Graph shows 2 theta/degree from 30 to 80 with plots for B equals 6T, P equals l006/(l006 plus l110) equals 0.54, et cetera.

Fig. 6 SiC slip with and without magnet oriented in magnetic field (6T)

Fig.6 is the XRD spectrum of two kinds of SiC ceramics fabricated at magnetic field. One is d⁰ ferromagnetic SiC with 1.0at% Al doping, the other is pure SiC powder (commercial). The two kinds slip with 30vol% solid content were poured into the plaster mold, then placed in a magnetic field of 6T for 2h. After drying, the greenbodies were sintered. Its grain orientation degree was characterized by XRD. It was found that in the plane perpendicular to the magnetic field, grains of pure SiC ceramics showed a random arrangement. Its orientation degree was 0.54, similar to samples without magnetic field treatment; while grains of d⁰ ferromagnetic SiC ceramic were partially oriented. Its 006 peak is obviously prominent, and the rest of the peak was significantly diminished. Its orientation degree was 0.87. According to Fig.5, the commercial SiC powder is also ferromagnetism, but it failed to make the grain oriented in 6T magnetic field. Relationship and mechanism between magnetic property of powder and its orientation at magnetic field remains to be studied.

Images show magnetic force of d0 ferromagnetic SiC ceramics labeled a, b, and c, and pure SiC ceramics labeled d, e, and f.

Fig. 7 Magnetic force microscopy images of d⁰ ferromagnetic SiC ceramics (a, b, c) and pure SiC ceramics (d, e, f). (a,d) surface morphology, (b, e) magnetic domain, (c. f) magnetic domain rotating 45o in the base of b).

Fig. 7 is the magnetic force microscopy (MFM) images of d⁰ ferromagnetic SiC ceramics (1.0at%Al) and pure SiC ceramics. According to our knowledge, the magnetic domain is first observed in the SiC ceramics. At room temperature, samples with d⁰ ferromagnetism show typical soft magnet strip domain (Fig.7(b)). In order to determine its real existence, samples were rotated 45o. Its strip domain is still clearly visible after rotation (Fig.7(c)). However, no magnetic domain shows at pure SiC ceramics.

CONCLUSIONS

Al:SiC ceramics show d⁰ ferromagnetism in the experiment. Strong magnetic hysteresis loop and magnetic domain in Al:SiC ceramics was observed for the first time. After slip casting in 6T magnetic field and sintering, SiC ceramics were well-oriented, and magnetic domain was detected. First-principles calculations were used to calculate spin magnetic moment of Al: SiC.

ACKNOWLEDGMENTS

Project supported by the National Natural Science Foundation of China (Grant Nos. 51572276) and Youth Innovation Promotion Association, CAS.

REFERENCES

Coey, J. M. D., 2005, d0 ferromagnetism, Solid State Sciences, 7(6), pp. 660-667.

Coey, J., Venkatesan, M., Fitzgerald, C., Douvalis, A., and Sanders, I., 2002, Ferromagnetism of a graphite nodule from the Canyon Diablo meteorite, Nature, 420(6912), pp. 156-159.

Venkatesan, M., Fitzgerald, C., and Coey, J., 2004, Thin films: unexpected magnetism in a dielectric oxide, Nature, 430(7000), pp. 630-630.

Coey, J., Venkatesan, M., and Fitzgerald, C., 2005, Donor impurity band exchange in dilute ferromagnetic oxides, Nature Materials, 4(2), pp. 173-179.

Zheng, H. W., Yan, Y. L., Lv, Z. C., Yang, S. W., Li, X. G., Liu, J. D., Ye, B. J., Peng, C. X., Diao, C. L., and Zhang, W. F., 2013, Room-temperature ferromagnetism in Cu-implanted 6H-SiC single crystal, Applied Physics Letters, 102, 142409.

Morbec, J. M., and Rahman, G., 2013, Role of vacancies in the magnetic and electronic properties of SiC nanoribbons: An ab initio study, Physical Review B, 87(11), 115428.

Wang, H., Yan, C. F., Kong, H. K., Chen, J. J., Xin, J., and Shi, E. W., 2012, Investigation of room temperature ferromagnetism of 3C-SiC by vanadium carbide doping, Applied Physics Letters, 101, 142404.

Liu, Y., Wang, G., Wang, S., Yang, J., Chen, L., Qin, X., Song, B., Wang, B., and Chen, X., 2011, Defect-induced magnetism in neutron irradiated 6H-SiC single crystals, Physical review letters, 106(8), p. 087205.

Suzuki, T. S., Uchikoshi, T., and Sakka, Y., 2006, Control of texture in alumina by colloidal processing in a strong magnetic field, Science and Technology of Advanced Materials, 7(4), pp. 356-364.

Zhu, X., Sakka, Y., Suzuki, T., Uchikoshi, T., and Kikkawa, S., 2010, The c-axis texturing of seeded Si3N4 with β-Si3N4 whiskers by slip casting in a rotating magnetic field, Acta Materialia, 58(1), pp. 146-161.

Mao, X. J., Wang, S. W., Shimai, S., and Guo, J. K., 2008, Transparent Polycrystalline Alumina Ceramics with Orientated Optical Axes, Journal of the American Ceramic Society, 91(10), pp. 3431-3433.

Furushima, R., Kato, Z., Uematsu, K., and Tanaka, S., 2013, Influence of Aggregates in α-Al2O3 Slurry on Orientation Degree of Powder Compact Fabricated by Magnetic Forming Method, Journal of the American Ceramic Society, 96(8), pp. 2411-2418.

Grasso, S., Hu, C. F., Vasylkiv, O., Suzuki, T. S., Guo, S. Q., Nishimura, T., and Sakka, Y., 2011, High-hardness B4C textured by a strong magnetic field technique, Scripta Materialia, 64(3), pp. 256-259.

Ni, D. W., Zhang, G. J., Kan, Y. M., and Sakka, Y., 2011, Textured h-BN Ceramics Prepared by Slip Casting, Journal of the American Ceramic Society, 94(5), pp. 1397-1404.

Sakka, Y., and Suzuki, T. S., 2005, Textured development of feeble magnetic ceramics by colloidal processing under high magnetic field, Journal of the Ceramic Society of Japan, 113(1313), pp. 26-36.

Innovative Processing and Manufacturing

CONTROLLING FACTORS FOR CREATING DENSE SiC-POLYCRYSTALLINE FIBER

Ryutaro Usukawa and Toshihiro Ishikawa

Tokyo University of Science, Yamaguchi

1-1-1 Daigaku-Dori, Sanyo-Onoda, Yamaguchi, 756-0884, Japan

ABSTRACT

A polymer-derived SiC-polycrystalline fiber, which was synthesized from an amorphous Si-Al-C-O fiber, shows excellent heat-resistance up to 2000oC with relatively high mechanical strength. This type of polymer-derived SiC-polycrystalline fiber is synthesized by a conversion process from the amorphous Si-Al-C-O fiber into SiC-polycrystalline fiber. In this conversion process, a degradation reaction of the amorphous Si-Al-C-O fiber and the subsequent sintering of the degraded fiber proceed. The degradation reaction is sensitively affected by the following factors: (a) Partial pressure of SiO-gas, (b) Partial pressure of CO-gas, (c) Heating rate and degradation temperature, (d) Argon gas flow. Accordingly, to obtain the desirable crystalline structure, we should strictly control the reaction conditions during the conversion process. In this paper, lots of changes, which occur during the degradation reaction of the amorphous Si-Al-C-O fiber and the subsequent sintering of the degraded fiber, will be described in detail, and also a new idea for obtaining the denser structure will be introduced.

INTRODUCTION

Among commercialized ceramic fibers, SiC fibers show relatively high mechanical strengths up to high temperatures over 1000oC in air [1-4]. Accordingly, active research and development on various composite materials using the SiC fibers have been conducted [5-7] and engine manufactures have actively evaluated the composite materials. The first developed SiCbased fibers have been produced in the middle 1960s by chemical vapor deposition (CVD process) onto tungsten or carbon filament core. However, as these types of SiC-based fiber had a large diameter, their applications were limited by their difficulties for use. After that, a SiC-based fiber with small diameter of about 10 micron meters was synthesized from organo-silicon polymer. This type of fiber was classified into a polymer-derived SiC fiber. The first polymer-derived SiC fiber was synthesized from polycarbosilane by Yajima and his coworkers [8] in the middle 1970s. After that, many types of polymer-derived SiC fibers have been developed and commercialized [3,4]. Of these polymer-derived SiC fibers, stoichiometric SiC-polycrystalline fibers have an excellent heat-resistance up to 2000oC [9-11]. Therefore, representative aircraft engine manufacturers are expecting actual applications of the SiC-polycrystalline fibers for jet engines and land-based gas turbines as the SiC/SiC composite materials. By the way, the mechanical strengths of the composite materials are dominated by the fiber’s strength. Hence, to extend the application field, increase in the mechanical strengths of these fibers is eagerly required. The strength of all polymer-derived SiC fibers is about 3GPa, which is remarkably low compared with the theoretical strength (about 40GPa) of SiC crystal. The mechanical strength of the fiber has been found to be strongly dominated by defects contained in each filament [1]. Previously, existence of some defects (for example: residual carbon, and so on) has been confirmed [12,13]. Accordingly, to increase the fiber’s strength, it’s very important to decrease the formation of the residual defects during the production process. Present polymer-derived SiC-polycrystalline fiber is synthesized via a conversion process from an amorphous structure into SiC-polycrystalline structure [9]. In this conversion process, a degradation reaction of the amorphous raw fiber (for example: Si-Al-C-O fiber) and the subsequent sintering of the degraded fiber proceed. The degradation reaction is sensitively affected by the following factors: (a) SiO-gas partial pressure, (b) CO-gas partial pressure, (c) Heating rate and degradation temperature, (d) Argon gas flow. Accordingly, to obtain the desirable crystalline structure, we should strictly control the reaction conditions during the conversion process. In this paper, the most important structural changes and controlling factors will described, and a new process for creating the desirable fine structure will appear.

EXPERIMENTAL

The SiC-polycrystalline fiber was synthesized by heat-treatment up to 1900oC of an amorphous Si-Al-C-O fiber, which is synthesized from polyaluminocarbosilane. The polyaluminocarbosilane was synthesized by a reaction of polycarbosilane with tetrabutoxyaluminum at 300oC in nitrogen atmosphere. A spun fiber was obtained by melt-spinning of the polyaluminocarbosilane, and then the spun fiber was cured at around 200oC in air. The cured fiber was fired at around 1300oC in nitrogen atmosphere to obtain the amorphous Si-Al-C-O fiber. The Si-Al-C-O fiber was composed of SiC fine crystals, oxide phases (estimated forms: SiO2, AlOx), and excess carbons. By the way, as mentioned above, since in this synthesis we used polyaluminocarbosilane which was synthesized by the reaction of polycarbosilane and tetrabutoxyaluminum, we presumed that the aluminum existed as some oxide forms in the Si-Al-C-O fiber. In the next step, the amorphous Si-Al-C-O fiber was heat-treated up to around 1500oC in argon gas atmosphere. During the heat-treatment, by the existence of the oxide phase and excess carbon in the fiber, the amorphous Si-Al-C-O fiber was degraded accompanied by a release of CO gas to obtain a porous degraded fiber. The porous degraded fiber was composed of a nearly stoichiometric SiC composition containing small amount of aluminum. In this case, since a part of the aluminum contained in the amorphous Si-Al-C-O fiber might be disappeared as some oxide materials during the heat-treatment process, consequently a very small amount of aluminum (less than 1wt%) was contained in the degraded fiber. By the existence of the small amount of aluminum, at the next step, an effective sintering proceeded in each degraded filament composed of the nearly stoichiometric SiC crystals during further heat-treatment up to 2000oC in argon atmosphere. The production scheme of the polymer-derived SiC polycrystalline fiber using the Si-Al-C-O fiber as the raw fiber is shown in Fig.1. In this research, structural changes during the conversion process from the Si-Al-C-O fiber into the SiC-polycrystalline fiber was addressed.

Diagram shows chemical structure where RO-Al-(OR)2 leads to polyaluminocarbosilane, which leads to amorphous structure Si-Al-C-O fiber, degraded fiber (porous structure), and beta-SiC polycrystalline fiber (dense structure).

Fig. 1 The production scheme of the SiC-polycrystalline fiber using a raw Si-Al-C-O fiber

Research on the heat-treatment (degradation reaction and sintering) of the Si-Al-C-O fiber was performed using Super High Temperature Inert Gas Furnace (NEWTONIAN Pascal-40, Produced by NAGANO) under argon gas flow (1 L/min). The size of the heating zone (made of graphite and C/C composites) is 35mm in diameter and 40mm in height. The programing rate and the maximum temperature were 400oC and 2000oC, respectively.

The surface and cross section of the obtained fibers were observed using a field emission scanning electron microscope (FE-SEM), model JSM-700F (JEOL, Ltd.). Parts of surface region and inside of the several samples were sharpened by an etching machine using focused ion beam (FIB), and then the fine structures were observed by the transmission electron microscope (TEM), model JEM-2100F (JEOL, Ltd.).

Auger electron spectroscopy (AES) depth profiles of Si, Al, C, and O were obtained using an ULVAC PHI SMART-200 operating at 3kV.

RESULTS AND DISCUSSION

During the conversion process from the amorphous Si-Al-C-O fiber into the SiCpolycrystalline fiber, lots of changes occur. The possible changes were summarized in Table 1. Of these changes, the most important one for obtaining SiC-stoichiometric composition is the following reaction. This reaction thermodynamically proceeds over 1522oC accompanied by a release of CO gas.

(1)

numbered Display Equation

However, the other changes simultaneously occur. Accordingly, we should strictly control all changes during the conversion process. Among these changes, especially we should beware the vaporization of SiO gas. Of course, the partial pressure of SiO gas is very important for causing another reaction (SiO+2C=SiC+CO). The most important thing is that these changes should effectively occur in the inside of each filament. However, the vaporization of SiO gas is usually easy to occur from inside to outside of each filament, and then a part of SiO gas is disappeared and some part of SiO gas react with excess carbon at the surface region of each filament to form an abnormal SiC crystal grain. Therefore, we should strictly control this vaporization of SiO gas.

Table 1. Possible changes during the conversion process from an amorphous Si-Al-C-O fiber into the SiC-polycrystalline fiber

Table shows columns for temperature in degrees Celsius, 1000, 1100, 1200, 1300, 1400, 1500, 1600, 1700, 1800, 1900, and 2000, and rows for main reaction, sub-reaction, vaporization of SiO gas, phase transformation of SiO2, active diffusion of excess carbon, and sintering of SiC crystals.

As can be seen from Table 1, during the heat-treatment process of Si-Al-C-O fiber the vaporization of SiO gas remarkably occurs over 1100oC, and at temperatures over 1500oC remarkable carbon-diffusion from inside to outside of each filament occurs. Accordingly, if the vaporization of SiO gas was not prevented, lots of abnormal SiC-crystalline grain growth would occur at the surface region (Fig.2). This is a large problem, because the surface roughness surely decreases the mechanical strength of the fiber.

Images show surface structure changes with markings for remarkable vaporization of SiO gas and remarkable carbon-diffusion, and labels for initial, 1000 degrees Celsius, 1100 degrees Celsius, 1200 degrees Celsius, 1300 degrees Celsius, et cetera.

Fig. 2 Changes in the surface structure after immediate heating up to various temperatures This type of surface grain growth was caused by the reaction between SiO gas and diffused carbon according to the following reaction.

(2)

numbered Display Equation

Furthermore, the SiO-disappearance by the vaporization of SiO gas consequently leads to an increase in the residual carbon as can be seen from the abovementioned equation (2). Fundamentally to obtain a stoichiometric SiC composition the above two reactions (equation (1) and (2)) should smoothly occur. However, as mentioned above, if the SiO-disappearance occurred by the vaporization, the excess carbon is consequently retained. The cross-section of the undesirable fiber obtained under the undesirable condition is shown in Fig.3.

Images show labels for cross-section of heat-treated fiber after immediate heating at 1900 degrees Celsius and polished cross-section, and markings for undesirable condition and SiO-disappearance occurred during initial degradation.

Fig. 3 The cross-section of the undesirable fiber containing lots of residual carbon

This figure shows the cross-section of the undesirable fiber. The right-side photograph shows the polished cross-section. The black points in the cross-section is the residual carbon, which was caused by the disappearance of SiO gas during the heat-treatment process. Accordingly, it is very important to control the vaporization of SiO gas from the inside of each filament.

Hence, to prevent the disappearance of SiO gas, we proposed a new process using self-forming and self-consuming "Just-in-time Reactor system. In this case, the Just-in-time Reactor" is an oxygen-rich surface layer which exists on each filament just only during the initial degradation process. This oxygen-rich surface layer is formed by in-situ process before the initial degradation, and after that the oxygen-rich surface layer should be smoothly disappeared. This new concept is shown in Fig.4. By the existence of oxygen-rich surface layer, it is expected that the vaporization of SiO gas from the inside of each filament is effectively prevented. It is because the partial pressure of SiO gas is also supplied from the oxygen-rich surface layer into the inside of each filament.

Flow diagram shows reactant mixture leads to self-formed surface layer composed of reactant A, which leads to uniform, stoichiometric composition, and reactant A plus reactant B leads to intermediate products, which leads to further physical changes and final product.

Fig. 4 A general process using self-forming and self-consuming "Just-in-time Reactor" system

Detailed process is shown in Fig. 5. This process contains a little bit difference from the production scheme shown in Fig. 1.

Diagram shows four cylinders with markings for partial oxidation by in-situ process, surface layer was disappeared, degradation, sintering, and oxygen-rich surface layer.

Fig. 5 Detailed process using the "Just-in-time Reactor" system

In this concrete process, to prevent the vaporization of SiO gas from the inside of each filament, partial oxidation by in-situ process was performed before the first degradation process. By the existence of the oxygen-rich surface layer, the vaporization of SiO gas was effectively prevented. And, after the initial degradation, at higher temperatures the oxygen-rich surface layer was smoothly disappeared. After that, the most important reaction (SiO2+3C→SiC+2CO) proceeds to create the stoichiometric SiC composition, and at higher temperatures the sintering in the inside of each filament proceeded to obtain dense structure.

The surface composition of the pre-oxidized raw fiber is shown in Fig.6. As can be seen from this figure, the thickness of the oxygen-rich surface layer of the pre-oxidized raw fiber was about 50nm. And during the initial degradation process, the important nucleation proceeds in the inside of each filament, that is, in each "Just-in-time Reactor".

Graph shows depth in nm from 0 to 200 versus atomic concentration in percentage from 0.0 to 80.0 where curves labeled Si and Al increase, decrease, and stay level, C increases and stays level, and O decreases and stays level.

Fig. 6 Auger electron spectroscopy (AES) depth profiles of the pre-oxidized raw fiber

As mentioned before, the nucleation of the SiC crystal is very important for creating the dense structure of the obtained SiC-polycrystalline fiber. And it was found that the nucleation proceeded at about 1400oC [14]. Therefore, the temperature region around 1400oC needs close attention. As mentioned before, the SiC nucleation proceeds by the following reaction.

SiO(g) + 2C → SiC + CO(g) (⊿G<0 at all temperature regions)

Therefore, the disappearance of SiO gas should be effectively prevented. Accordingly, our new proposal using "just-in-time Reactor" system plays an important role as a protective layer for preventing the disappearance of SiO gas. Fig.7 shows the TEM image of the surface region of the pre-oxidized raw fiber after immediate heat-treatment at 1410oC in argon atmosphere.

Images show surface region with markings for SEM, TEM image, surface oxide layer, core SiC, SiC(1 nm), and SiC(4 nm), and table shows columns for measurement and literature, and rows for a, b, and c.

Fig. 7 TEM image of the surface region of the pre-oxidized raw fiber after immediate heat-treatment at 1410oC in argon atmosphere.

As can be seen from Fig.7, the surface oxide layer effectively remained, and in the inside of the fiber the SiC-nucleation (1~4nm) was observed. And, it was found that these crystals were cubic SiC fine crystals.

Fig. 8 shows the TEM image of the surface region of the pre-oxidized raw fiber after immediate heat-treatment at 1450oC in argon atmosphere. In this case, the surface oxide layer was perfectly disappeared and then the crystalline grain growth uniformly proceeded. By the use of the pre-oxidized raw fiber, abnormal crystalline growth was effectively prevented.

Images show surface region with markings for SEM, TEM image, surface, core SiC, SiC(11 nm) and SiC(24 nm).

Fig. 8 TEM image of the surface region of the pre-oxidized raw fiber after immediate heat-treatment at 1450oC in argon atmosphere.

Fig. 9 shows the cross-section of the SiC-polycrystalline fiber synthesized by the use of pre-oxidized Si-Al-C-O fiber along with the comparative result using non-oxidized fiber. As can be seen from Fig.9, by the use of the pre-oxidized Si-Al-C-O fiber (our "Just-in-time Reactor" system), the residual carbon was remarkably reduced.

Flow diagram shows SiC-polycrystalline fiber synthesized using non-oxidized Si-Al-C-O fiber leads to (remarkably improved) SiC-polycrystalline fiber synthesized using oxidized Si-Al-C-O fiber.

Fig. 9 Cross-section of the SiC-polycrystalline fiber synthesized by the use of pre-oxidized Si-Al-C-O fiber along with the comparative result using non-oxidized fiber.

CONCLUSIONS

In order to produce a stoichiometric dense structure of SiC-polycrystalline fiber using an amorphous Si-Al-C-O fiber as a starting material, we should strictly control the degradation and sintering processes. Especially, the degradation process is very important to prevent an abnormal grain growth of SiC crystals and to reduce the residual carbon. In this research, it was found that the most important thing is to prevent the vaporization (disappearance) of SiO gas from the inside of each filament during the degradation process. To prevent the vaporization of SiO gas, we proposed a new "just-in-time Reactor system, which played an important role as a protective layer. The oxygen-rich surface layer effectively existed during only initial degradation process, and after a SiC-nucleation uniformly proceeded in the inside of each pre-oxidized Si-Al-C-O fiber (just-in-time Reactor"), the surface oxide layer was effectively disappeared to create the desirable dense structure with lesser residual carbon.

REFERENCES

T.Ishikawa and H.Oda, Defect control of SiC polycrystalline fiber synthesized from polyaluminocarbosilane, J.Euro.Ceram.Soc., 36 (2016) 3657-3662.

M.Wilson and E.Opila, A review of SiC fiber oxidation with a new study of Hi-Nicalon SiC fiber oxidation, Advanced Engineering Materials, DOI: 10.1002/adem.201600166 (2016) 1-12.

O.Flores, R.K.Bordia, D.Nestler, W.Krenkel, and G.Motz, Ceramic Fibers Based on SiC and SiCN Systems: Current Research, Development, and Commercial Status, Advanced Engineering Materials, 16(6) (2014) 621-636.

P.Colombo, G.Mera, R.Riedel, and G.D.Soraru, Polymer-Derived Ceramics: 40 Years of Research and Innovation in Advanced Ceramics, Ceramic Science and Technology: Volume 4: Applications Edited by Ralf Riedel and I-Wei Chen, (2013) 245-320.

J.J.Sha, T.Nozawa, J.S.Park, Y.Katoh, and A.Kohyama, Effect of heat treatment on the tensile strength and creep resistance of advanced SiC fibers, Journal of Nuclear Materials, 329-333 (2004) 592-596.

K.Itatani, K.Hattori, D.Harima, M.Aizawa, and I.Okada, Mechanical and thermal properties of silicon-carbide composites fabricated with short Tyranno Si-Zr-C-O fiber, Journal of Materials Science, 36 (2001) 3679-3686.

N.Remirez de Esparza, N.Cocera, L.Vazquez, J.Alkorta, I.Ocana, and J.M.Sanchez, Characterization of CVD Bonded Tyranno Fibers Oxidized at High Temperaturs, J.Am.Ceram.Soc., 97[12] (2014) 3958-3966.

S.Yajima, M.Omori, J.Hayashi, K.Okamura, T.Matsuzawa, and C.Liaw, Symple synthesis of the continuous SiC fiber with high tensile strength, Chem.Lett., (1976) 551-554.

T.Ishikawa, Y.Kohtoku, K.Kumagawa, T.Yamamura, and T.Nagasawa, High-strength alkali-resistant sintered SiC fibre stable to 2200oC, Nature, 391 (1998) 773-775.

M.Takeda, A.Urano, J.Sakamoto, and Y.Imai, Microstructure and oxidative degradation behavior of silicon carbide fiber Hi-Nicalon type S, Journal of Nuclear Materials, 258-263 (1998) 1594-1599.

T.Ishikawa, Advances in Inorganic Fibers, Advanced Polymer Science (Springer-Vrlag Berlin Heidelberg) 178 (2005) 109-144.

C.Sauder, and J.Lamon, Tensile Creep Behavior of SiC-Based Fibers with a Low Oxygen Content, J.Am.Ceram.Soc., 90(4) (2007) 1146-1156.

J.J.Sha, J.S.Park, T.Hinoki, and A.Kohyama, Tensile behavior and microstructural characterization of SiC fibers under loading, Materials Science and Engineering A, 456 (2007) 72-77.

R.Usukawa, H.Oda, and T.Ishikawa, Conversion process of amorphous Si-Al-C-O fiber into nearly stoichiometric SiC polycrystalline fiber, Journal of the Korean Ceramic Society, 53(6) (2016) 610-614.

ECO-FRIENDLY SYNTHESIS OF GRAPHENE USING HIGH PRESSURE AIRLESS SPRAY SYSTEM

Karanveer S. Aneja ¹, Anand Khanna ¹*

¹ Department of Metallurgical Engineering and Materials Science,

 Indian Institute of Technology Bombay, Mumbai 400076, India.

*E-mail: [e-mail]

ABSTRACT

Graphene appears to be a wonder material of the present decade. Its special properties such as electrical conductivity, electronic mobility, mechanical strength, optical properties have caught the attention of both the scientific as well as industrial community. One of the biggest bottlenecks of its widespread usage is its availability in large volumes. Hence any research which focuses on industrial scalable production is the need of the hour. Keeping this in mind, a novel high pressure eco-friendly approach has been developed to produce graphene that can cater to the industrial demands of graphene.

1. INTRODUCTION

Graphene, also termed as the wonder material, has generated a lot of curiosity in the scientific and industrial community since its invention in 2004 [1]. Graphene is a two dimensional structure consisting of a single layer of carbon atoms. With a tensile strength of 130GPa [2], Graphene is the strongest material ever discovered. It has extremely high electrical conductivity and electron mobility [3]. Any application of graphene will find it difficult to see the light of the day unless it is available in plenty from an ecofriendly synthesis method.

There have been multiple graphene producing routes which either use Chemical Vapor Deposition or chemical and mechanical energy to exfoliate graphite into graphene. Unfortunately, these routes suffer from high cost, low yield or a combination of these factors. Hence, developing an alternate route of graphene production was warranted that can help cater to the large volumes of defect free graphene requirement.

High pressure exfoliation or Airless spray pressure technique was devised after taking into account the drawbacks of the available graphene production routes. The use of pressure can be explained using a simple example of a stack of paper. When one blows a stack of paper, it separates out into individual sheets. Similarly, graphite can be thought as a stack of individual graphene sheets. If it subjected to air pressure, using a spray machine for example, it should separate out into individual layers.

Diagram shows air blow and stack of paper leads to set of diagonal lines, and diagram shows air pressure and graphite leads to graphene.

Figure 1: Exfoliation using air pressure

Although Vander Wall forces of attraction, that hold the individual graphene sheets together, are weak in nature, it requires significant force to overcome this binding force. The use of supersonic acceleration of droplets through a Laval nozzle has been commonly used to exfoliate graphene [6-8]. The limitation of using the supersonic nozzle is the damage it causes to the structure