Fabrication and Microstructural Investigations of Porous Ceramic Particle/Polymer Matrix Composites

By Sharif, M A Sueyoshi, H

Well defined porous ceramic particle/polymer matrix composites consisting of Si and nanosized ZrO^sub 2^ particles have been fabricated by the pyrolysis of phenolic resin at 1123 K in vacuum. Electron probe microanalyser analysis showed that the pyrolysis of the starting materials led to the agglomeration of ZrO^sub 2^ particles and the uniformly distribution of Si particles in the matrix. Scanning electron microscopy images showed the formation of spherical pores ranging from 10 [mu]m to several 100 [mu]m in diameter in the composite, resulting in the flexural strength of ~14 MPa. X-ray diffraction analysis suggested that beta-SiC might be formed in the composite; and the matrix of porous polymer matrix ceramic composite was amorphous carbon. Keywords: Pyrolysis, Composites, EPMA, Microstructure, beta-SiC

Introduction

Porous ceramic particle/polymer matrix composites are nowadays commonly researched and developed for variety of applications. For example, particles of SiO^sub 2^, SiC or Zr02 are added to polymer matrix to improve its properties such as hardness, stiffness and wear resistance.1 Polymer and polymer matrix composites have been recognised as great potentials in industry as a class of triboengineering materials.2 Porous materials are of significant interest due to their wide applications in catalysis, separation, light weight structural materials, biomaterials, etc.3 Various processing techniques have been employed to tailor different porous structures in ceramic materials. The use of pore former (polymer) that evolves gases during pyrolysis is a common method. Partial sintering has also been used as an alternative approach for obtaining the porous structures. To date, a variety of materials with porous structures have been successfully fabricated, such as zirconia, alumina, silicon carbide and titanium oxide. However, the strength of porous ceramics is considerably low compared to dense ceramics, and one of the most significant developments required to porous ceramics is to improve their mechanical properties.

One of the methods to strengthen porous ceramics is to fabricate strongly bonded matrix such as SiC between pores. In addition, the introduction of small pores into the matrix is favoured because fracture strength of brittle materials decreases with increasing flaw size.4-6 Another effective method for strengthening porous ceramics is to reinforce the matrix by incorporating second particles or fibres into the matrix.7’8 In order to attain these strengthening effects simultaneously during the synthesis of porous ceramics, it is necessary to prepare preforms of ceramic matrix composites (CMCs) with matrix in which fine pores and second particles are uniformly distributed.

Phenolic resins are known for their high temperature resistance and high char yielding properties.9 Polymeric materials can be fluidised and decomposed during the heat treatment of pyrolysis. The decomposition of polymeric materials is associated with the generation of a variety of gases, which in turn make pores into the obtained matrix. The addition of silicon into carbon leads to the formation of SiC when the heating temperature is >1673 K.10 Furthermore, it is expected that the porous SiC matrix may be toughened by incorporating hard particles into the matrix.

Zirconia is a well known material which has superior properties such as high mechanical strength, chemical durability, alkali resistance, heat resistance against oxidation and refractoriness.11 The importance of zirconia in various technical applications makes the compound an interesting subject for research work. Over the past years, there has been an increasing interest in nanostructured ceramics for their lower sintering temperature and improved mechanical properties.12

Therefore, to fabricate porous ceramic particle/ polymer matrix composites, phenolic resin was mixed with Si and ZrO^sub 2^ powders, and then heated at 1123 K in a closed copper tube.

Experimental

Starting materials

The average grain sizes of solid Novolac type phenolic resin (PR- 50590B, Sumitomo Bakelite Co., Ltd. Japan), Si (Furuuchi Chemical Co., Ltd. Japan) and ZrO^sub 2^ (Tosoh-zirconia TZ-O, Tosoh Co., Ltd. Japan) powders are 35 [mu]m (max. 40 [mu]m, min. 20 [mu]m), 0.40 [mu]m (max. 0.68 [mu]m, min. 0-23 [mu]m) and 1 [mu]m (max. 1.13 [mu]m, min. 0.44 [mu]m) respectively. In order to maintain 1:1 atomic ratio between silicon and carbon atoms (for the formation of silicon carbide) in the composites, the calculated mass ratio between phenolic resin and silicon is1 : 1.5. But considering the loss of carbon atoms due to the formation of gaseous molecules during pyrolysis, the designed mass ratio of phenol resin is larger than the calculated value. The mass ratio of ZrO^sub 2^, Si and phenol resin is 1 : 1 : 2 respectively.

Sample preparation

The powders of phenol resin, ZrO^sub 2^ and Si were mixed in a grinding bowl for 3 h, and then mixture of powders was put into a copper tube whose both ends were closed by mechanical pressing. Pyrolysis was carried out by heating the copper tube at 1123 K in a vacuum furnace.

Properties and morphology investigation

The weight loss and apparent density of the pyrolysed specimens were measured. Macroscopic appearance of the pyrolysed specimen was observed using charge coupled device (CCD) camera (digital camera, CP 4500, Nikon Co. Ltd., Japan), and the microstructure of the polished specimens was observed using scanning electron microscopy (SEM; XL-30 ESEM Series, FEI Co. Ltd., Japan). The energy dispersive spectrometer (EDS), equipped to the SEM, was used for the elemental analyses. The distributions of elements involved in the pyrolysed specimens were examined by using an electron probe microanalyser (EPMA, TXA-8600SX, JEOL Co. Ltd., Japan).The crystalline phases were determined by X-ray diffraction (XRD-6000S Shimadzu Co. Ltd., Japan), which was performed with Cu irradiation (1.54060 A) at a scanning rate of 2[degrees] min^sup -1^ in 2theta range between 20 and 80[degrees]. In addition, flexural strengths of the specimens were measured at room temperature using a three point bending test fixture (span: 30 mm, crosshead speed: 0.5 mm min^sup -1^), which was mounted on universal material testing machine (AG-1, 50kN, Shimadzu Co. Ltd., Japan).

Results and discussion

Weight loss and density

The average weight loss in the specimens was 19.70% (max. 20.20%, min. 19.00%), which was attributed to the degasification of molecules formed during pyrolysis. The average apparent density of specimens was 0-85 g cm-3 (max. 1.01 g cm^sup -3^, min. 0.68 g cm^sup -3^). This lightweight composite is devoted to the formation of many pores.

Macrostructure

Figure 1 shows the appearance of porous ceramic particle/polymer matrix composite pyrolysed at 1123 K. The outer surface of the composite is smooth, while the cross-sectional observation reveals the existence of pores. The specimen has a diameter of ~7 mm, which is 1 mm less than the inner diameter of copper tube used as the heating mould. These results suggest that pores are formed during heating, which causes the mixtures to dilate within the mould.

1 Appearance of pyrolysed composite

2 Image (SEM) showing cross-sectional view of phenol resin- Zr0^sub 2^-Si composites

Microstructure

Figure 2 shows the cross-sectional SEM image of pyrolysed composite containing spherical pores of size ranging from several 10 [mu]m to several 100 [mu]m in diameter, some of which are interconnected with each other. The voids resulted from the degradation of polymer components.13 Particles of several 10 [mu]m in diameter exist in the matrix, which appear as relatively white images (Fig. 1). It is obvious from these results that the small average density of the composites resulted due to the existence of uniformly distributed pores in the interior of the composites.

Figure 3 is an enlarged view of a particle (relatively white image) in the matrix of the composite and seems to have layer structure: one is the outer layer (point B) with a thickness of ~10 [mu]m, and the other is the interior (point A) surrounded by the outer layer.

In order to obtain the further information of materials, which constituted the outer layers (point B), interior (point A) of the particles and the surrounding matrix (point C), the detection of elements was carried out at points A, B and C (Fig. 3) respectively.

In the interior of the particle (point A in Fig. 3), the peak of Zr is the higher than those of C, O and Si, as shown in Fig. Aa. This result suggests that the interior of particles are not only made of nanoZrO^sub 2^ particles, but also contain C, O and Si atoms. Furthermore, low peak of C suggests the low content of C atoms compared to those of Zr, Si and O atoms.

3 Enlarged SEM image of matrix particle

4 Analyses (EDS) for detection of elements in a interior of particle (point A in Fig. 3), b outer layer of particle (point B in Fig. 3) and c matrix (point C in Fig. 3) of pyrolysed composite

In the outer layer of the particle (point B in Fig. 3), the peak height of Zr and O slightly decreases while that of C and Si increases, as shown in Fig. Ab. Therefore, in comparison to interior, the outer layer of the particle is also composed of Zr as a major element with more Si and C atoms. In the matrix (point C in Fig. 3), the peak of C is higher than those of Zr, Si and O (Fig. Ac). In comparison to interior and outer layer of the particles, matrix is composed of C as a major element with more Si and less Zr and O elements. Therefore, EDS analyses suggest that small particles were formed by the agglomeration of nano-Zr02 particles in powder mixing; and then these small particles were grown to large particles with larger amounts of Si and C in the outer layer. In addition, some reactions took place during the formation of particles (with inner white and outer greyish layer, Fig. 3).

5 Enlarged SEM image of cross-section of pyrolysed composite

Figure 5 is an enlarged SEM image of the cross-section of the pyrolysed composite, which shows the formation of particles in the matrix (particle D) and pores (particle E). section F shows the distribution of particles in a specific area of the matrix of the composite.

Figure 6 is a typical example of EMPA analyses of the section F (Fig. 5) in which the area distributions of Zr and Si were examined. Figure 6a shows that the particle is rich in Zr. Thus, in addition to EDS analyses, EPMA analysis also suggested that the particles were formed by the agglomerate of nano-ZrO^sub 2^ particles. Figure 6b shows that there are some spots and regions where Si is rich. Such spots and regions are located in the matrix surrounding the Zr rich particles. Thus, these spots are considered to be Si particles.

6 Images (EPMA) indicating area distributions of a Zr and b Si in phenol Tesin-ZrO^sub 2^-Si composite

7 X-ray diffraction spectrum obtained from pyrolysed composite

8 Load deflection curve obtained for pyrolysed composite

Analysis of crystalline structure

X-ray diffraction spectrum obtained from pyrolysed porous ceramic particle/polymer matrix composite is shown in Fig. 7. Monoclinic ZrO^sub 2^ has peaks (-111), (111), (200) and (002) at 2theta=28.1, 31.4, 34.1 and 35.3[degrees] respectively. The peaks at 2theta=28.4, 47.3 and 56.1[degrees] belong to (111), (220) and (311) planes of cubic crystalline silicon.14 Other peaks of Si and ZrO^sub 2^ are also indexed in Fig. 7. The low intensity XRD peaks around 2theta=35.6, 41.4, 60.0 and 71.8[degrees] are (111), (200), (220) and (311) respectively. These peaks belong to cubic form silicon carbide (beta-SiC).15,16 Amorphous carbon exists in the pyrolysed composite, because the peaks of carbon were not observed in XRD profile.

Three point bending test

Load deflection curve obtained from the three point bending test of pyrolysed specimen is shown in Fig. 8. The load increases linearly with deflection and at the maximum load specimens broke in a brittle manner. According to the theory of bending of beams, the maximum bending stresssigma^sub max^, which appears at the loading point, is used to evaluate the fracture strength of the composites. The maximum fracture stress that pyrolysed composite sustained was 14 MPa. Such low fracture strength may be attributed to pores existing in the specimens.

Conclusions

Porous ceramic particle/polymer matrix composites with well defined pore architecture were fabricated with the pore size ranging from 10 [mu]m to several 100 [mu]m in diameter. The low average density (0.851 g cm^sup -3^) of the pyrolysed composites is attributed to the formation of pores due to the degasification during pyrolysis. The EDS and EPMA analyses suggested that Zr is rich in the particles and there are some regions in the matrix surrounding the Zr rich particles where Si is rich. The EDS analyses also showed that the matrix of composite was composed of carbon. In addition, particles were nucleated due to the agglomeration of nanosized zirconia particles. X-ray diffraction analysis of the pyrolysed composite suggested the existence of beta-SiC. The maximum fracture strength of the pyrolysed specimen was at most 14 MPa. The experimental results have shown that this technique might be developed into a general pathway to prepare various porous composites with well controlled pore structure.

Acknowledgement

The authors would like to thank sincerely Mr Oozono Yosihisa (from Division of Instrumental Analysis, Frontier Science Research Center, Kagoshima University) for his assistance in conducting EPMA analyses.

References

1. K. Konopka, A. Boczkowska, K. Batorski, M. Szafran and K. J. Kurzydlowsk: Mater. Lett., 2004, 58, 3857-3862.

2. F.-H. Su, Z.-Z. Zhang and W.-M. Liu: Wear, 2006, 260, 861- 868.

3. Y. Sakka, F. Tang, H. Fudouzi and T. Uchikoshi: Sci. Technol. Adv. Mater., 2005, 6, 915-920.

4. S. M. Zhu, S. Q. Ding, H. A. Xi, Q. Li and R. D. Wang: Ceram. Int., 2007, 33, 115-118.

5. V. D. Krstic: Theor. Appl. Fract. Mech., 2006, 45, 212-226.

6. W. D. Kingery, H. K. Bowen and D. R. Uhlmann: ‘Introduction to ceramics’, 2nd edn, 791; 1960, New York, John Wiley.

7. K. K. Chawla: ‘Ceramic matrix composites’, 1st edn, 39; 1993, London, Chapman & Hall.

8. G. Ziegler, I. Richter and D. Suttor: Composites, 1999, 30, 411-417.

9. C. P. Reghunadhan Nair, R. L. Bindu and K. N. Ninan: Polym. Degrad. Stabil., 2001, 73, 251-57.

10. A. Ramond Cutler and M. Kevin Rigtrup: J. Am. Ceram. Soc, 1992, 75, 36-43.

11. J. I. Pena, H. Miao, R. I. Merino, G. F. de la Fuente and V. M. Orera: Solid State Ionics, 1997, 101, 143-147.

12. C. Jagadish Ray, K. Ranjan Pati and P. Pramanik: J. Eur. Ceram. Soc, 2000, 20, 1289-1295.

13. C-C. Ma, J.-M. Lin, W.-C. Chang and T.-H. Ko: Carbon, 2002, 40, 977-984.

14. G. D. Soraru, S. Modena, P. Bettotti, G. Das, G. Mariotto and L. Pavesi: Appl. Phys. Lett., 2003, 83, 749-751.

15. Z. C. Liu, W. H. Shen, W. B. Bu, H. R. Chen, Z. L. Hua, L. X. Zhang, L. Li, J. L. Shi and S. H. Tan: Microp. Mesop. Mater., 2005, 82, 137-145.

16. G-Q. Jin and X.-Y. Guo: Microp. Mesop. Mater., 2003, 60, 207- 212.

17. M. A. Schiavon, E. Radovanic and I. V. P. Yoshida: Powder Technol, 2002, 123, 232-241.

M. A. Sharif* and H. Sueyoshi

Graduate School of Science and Engineering, Kagoshima University, Kagoshima, 890 0065, Japan

* Corresponding author, email akhtarsharif@hotmail.com

(c) 2008 Institute of Materials, Minerals and Mining

Published by Maney on behalf of the Institute

Received 10 April 2007; accepted 2 August 2007

DOI 10.1179/174328407X241098

Copyright Institute of Materials Jan 2008

(c) 2008 Materials Science and Technology; MST. Provided by ProQuest Information and Learning. All rights Reserved.