Influence of Additives and Hot-Press Sintering on Mechanical and Lipophilic Properties of Silicon Nitride Ceramics*

Transcription

1 Materials Transactions, Vol. 46, No. 9 (25) pp. 241 to 246 #25 The Japan Institute of Metals Influence of Additives and Hot-Press Sintering on Mechanical and Lipophilic Properties of Silicon Nitride Ceramics* Mitsuo Kido, Tarou Tokuda, Rongguang Wang and Fumihiro Suzumura Department of Mechanical Systems Engineering, Faculty of Technology, Hiroshima Institute of Technology, Saeki-ku, Hiroshima , Japan Fe 3 O 4 and Mo and were added to hot-press sintered Si 3 N 4 ceramics to improve their lipophilic and mechanical properties. The bending strength, relative density, hardness and fracture toughness of Si 3 N 4 ceramics with added Fe 3 O 4 and Mo were improved by hot-press sintering compared with samples prepared with the pressureless process. In particular, the bending strength and relative density were improved by about 2 and 14%, respectively. Both the macro- and micro-lipophilicity of Fe 3 O 4 - and Mo-added ceramics were improved when prepared with the hot-press process, compared with those prepared with the pressureless process. This can be attributed to the addition of Fe 3 O 4, and the formation of MoO 3 and Si 2 N 2 O during hot-press sintering. The lower friction coefficient and high wear resistance of Fe 3 O 4 - and Mo-added ceramics have been achieved by applying the hot press process. The high wear resistance is considered to be due to the improvement of hardness and fracture toughness, which decreased the loss of particles during the wear test. (Received April 6, 25; Accepted July 2, 25; Published September 15, 25) Keywords: silicon nitride ceramics, iron oxide, molybdenum, hot pressing, mechanical property, lipophilic property, oil droplet, atomic force microscope, friction coefficient, wear volume 1. Introduction Recently, silicon nitride (Si 3 N 4 ) ceramics with iron oxide (Fe 3 O 4 ) and molybdenum (Mo) additives have attracted much attention because of their possible application as sliding parts with high mechanical strength, self-lubricity and lipophilicity. 1) It has been reported that the high lipophilicity came from the lipophilic Fe 3 O 4 component and the high selflubricity came from the layer-structured MoO 3 that is formed during sintering. 1,2) On the other hand, the authors found that the bending strength, hardness and density of such ceramics prepared by pressureless sintering decreased with the addition of Fe 3 O 4 and Mo, which was explained by a decrease in density and the existence of micro cracks in the ceramics. 2) Many reports have indicated that the hot-press sintering process can generally increase the density and strength of ceramics. 3,4) However, its effect on silicon nitride with significant quantities of added Fe 3 O 4 and Mo has not been reported. In the present study, the hot-press sintering process was applied to ceramic specimens prepared with added Fe 3 O 4 and Mo, and their mechanical properties, including bending strength, hardness, fracture toughness, friction coefficient, wear volume and their lipophilicity to macroand micro-droplets of lubricating oil were investigated. 2. Experimental Procedures 2.1 Specimen preparation, mechanical and lipophilic properties Si 3 N 4 powder (purity: 99.99%; average diameter:.4 mm) with added Al 2 O 3 powder (2.83 mass%) and Y 2 O 3 powder (2.83 mass%) was used to prepare Si 3 N 4 ceramics. Such specimens were designated. Various amounts of Fe 3 O 4 and Mo powders were also added to prepare other specimens. *This Paper was Originally Published in Japanese in J. Japan Inst. Metals 68 (24) These specimens, with a constant amount of Mo and variable amount of Fe 3 O 4, were designated FM a,fm 1b and FM 1c with increasing Fe 3 O 4 content, while specimens with constant Fe 3 O 4 and variable Mo content were designated FM 2b and FM 2c with increasing Mo. Specimens were prepared by pressureless and hot-press sintering (2 MPa, 273 K, 12.4 ks). The subscript HP indicates the hot-press process (example: ). All specimens were ground (apparatus: Okamoto Co., Ltd., PFG-45DX) and polished with emery paper (#1). Furthermore, FM and specimens were heat treated, with the aim of forming the solid lubricating phase of MoO 3. 2) Finally, all specimens were ultrasonically cleaned in acetone and dried with a warm airflow prior to use. The bending strength, relative density, Vickers hardness, fracture toughness, and friction coefficient of the ceramics was investigated, as well as the contact angle of lubricating oil droplets. Details have appeared in our earlier reports. 2,5) In the present work, millimeter-scale droplets of lubricating oil were considered to be macro-droplets, while nano-size droplets were designated micro-droplets. 2.2 Composition analysis The surface composition was analyzed with an electron probe micro analyzer (JOEL Co., JXA-89R: EPMA), an X- ray photoelectron spectroscope (XPS: Shimadzu Co., AXIS ULTRA) and an X-ray diffraction analyzer (XRD: Mac Science Co. Ltd., M21X). The analysis conditions for EPMA and XPS were the same as reported previously. 2) The target used in XRD analysis was Cu-K, the voltage and current were 15 kv and 1 ma. 3. Results 3.1 Mechanical and macro-lipophilic properties Table 1 shows the mechanical and macro-lipophilic properties of,, and (Fe 3 O 4 : 3.6 mass%, Mo:.9 mass%) specimens. All values are averaged ones, i.e., the average of 1 measurements of bending strength, 3

2 242 M. Kido, T. Tokuda, R. Wang and F. Suzumura Bending strength b [MPa] Table 1 Properties of,, and. Relative density [%] Vickers hardness Hv (P ¼ 98[N]) Fracture toughness Kc [MPam 1=2 ] Contact angle [ ] Friction coeficient of Vickers hardness, 3 of fracture toughness, 3 of density, 5 of the friction coefficient and 5 of the macro contact angle. It is clear that the mechanical and macro-lipophilic properties of Si 3 N 4 ceramics, whether Fe 3 O 4 and Mo were added or not, were improved by the hot-press process as compared with the pressureless process. In particular, the bending strength of hot-press sintered ceramics was significantly increased to about double that of pressureless sintered ceramics, and the relative density was increased to about 14% of samples prepared by the pressureless process. 3.2 Micro-lipophilic property According to earlier reports, the micro-wetting contact angle varies with the size of micro-droplets. 2,5) Although the macro-contact angle of oil on the FM specimen was lower with that on the specimen, the contact angles of microdroplets on both specimens were almost the same. 2) The reason was considered to be the large variation in size of the micro-droplets of oil (radius: 5 3 nm). 2,5) In the present study, micro-droplets of radius 8 1 nm were selected from among many others to evaluate the micro-lipophilicity of each surface. Figure 1 shows the contact angles of micro-droplets of lubricating oil, which were obtained by observation with the ac non-contact mode of an atomic force microscope (AFM). It is clear that the contact angles of micro-droplets of oil on the FM specimen decreased with increasing amounts of added Fe 3 O 4. In addition, the contact angles on hot-press sintered specimens were improved compared with those on the pressureless prepared specimen. Furthermore, the micro-lipophilicity of the specimen was significantly improved, compared with that of the FM specimen. In addition, the contact angle of micro-droplets of oil was somewhat lower than that of macro-droplets. 4. Discussion 4.1 Microstructure obtained by hot-press sintering The microstructures of each specimen were observed in order to clarify the mechanism of the improvement of mechanical properties (especially strength), the decrease of the contact angle of oil and the reduction of the coefficient of friction by hot-press sintering. Figure 2 shows the typical microstructure of FM and specimens as observed by scanning electron microscopy (SEM). Table 2 shows the porosity and the average diameter of pores. The diameter of a pore was measured both in the horizontal and vertical directions. It is clear that both the porosity and the average diameter of pores in specimens were smaller that those mic/ degree θ Micro Contact Angle, Radius of Oil Droplet, r/nm 8 r 1 FM FM1b FM1c Fe 3 O 4 (mass%) Fig. 1 Change of micro-contact angles of oil droplet on, FM, and. in FM specimens. The authors have reported that the strength of Si 3 N 4 ceramics decreased with the addition of Fe 3 O 4 and Mo. 2) According to Fig. 2, since the increase of porosity increases the number of places where [high] stress intensity originates, part of the improvement of strength and hardness of hot-press specimens (Table 1) can be attributed to the decrease of fracture origination spots, i.e., the decrease in number and size of pores. In addition, columnar grains appeared in the hot-press sintered specimen [Fig. 2(b)]. Figure 3 shows the diameters and aspect ratios of such rods. The diameter D is obtained from the short edge of the rod, and the aspect ratio is the ratio of the lengths of the long and short edges. Since the aspect ratio became larger after hotpress sintering, the improvement of fracture toughness of HP sintered specimens (Table 1) was considered to be related to the formation of rods with larger aspect ratio. 3,4,6) The improvement of hardness should be attributed to the decrease in porosity. 7) Figure 4 shows the wear volume of FM and specimens after the wear test. The wear volume of the FM specimen decreased with increasing Mo, which is attributed 4 6

3 Influence of Additives and Hot-Press Sintering on Mechanical and Lipophilic Properties of Silicon Nitride Ceramics 243 (a) Material FM (a) 3 D / µ m Grain Size, Material (b) Material (b) 15 Aspect ratio 1 5 Material Fig. 2 Surface morphology of FM and observed by SEM (Arrows are pore and columnar grain). Fig. 3 Grain size and aspect ratio of,, and. Table 2 Porosity and diameter of pore of Material,, and. Material Porosity, % Diameter of blow hole, d/mm to the production of solid lubricant MoO 3 in the specimen. 2) The MoO 3 decreases the friction coefficient during the wear test (Table 1). The wear volume of the specimen also decreased, compared not only with the specimen but also the FM a,fm 2b,FM 2c and specimens. This means that both the addition of components and the hot-press sintering process can increase the wear resistance of Si 3 N 4 ceramics. Figure 5 shows a typical wear trace for FM a and specimens after a wear test. When wear occurs from contact between Si 3 N 4 and alumina ceramics, particles from the soft Si 3 N 4 ceramics are usually broken out to form holes in the material. 8,9) According to Fig. 5, the number of large holes along wear traces in material is less than that in FM a material. Such holes in are smaller than those in FM (Fig. 2 and Table 2). These holes must be produced by the loss of particles during the wear test. In general, brittle ceramic materials with high hardness and fracture toughness Specific Wear Volume, V /( 1 8 m 2 /N) resist wear. The relationship among these variables can be expressed as follows. 8) W ¼ FM Mo (mass%) E 7=8 FM2b FM2c Kc 1=2 Hv 3=2 P9=8 x 1 3 Fig. 4 Change of Specific Wear Volume in, FM, and.

4 244 M. Kido, T. Tokuda, R. Wang and F. Suzumura (a) Material FM (a) (b) Material (b) Oil droplet Fig. 5 Wear scars on FM and observed by SEM. Here, W is the wear volume, is a constant, E is the Young s modulus, Kc is the fracture toughness, Hv is the Vickers hardness, P is the load, x is the wear distance. Thus, the improvement in wear resistance of the specimen can be attributed to the difficulty of losing particles due to improved fracture toughness and decreased friction coefficient. 4.2 Composition analysis In a previous paper, 2) the authors reported that the lipophilicity of ceramics with respect to millimeter-scale oil droplets was influenced not only by the addition of Mo and Fe 3 O 4 but also the residence of oil droplets in pores. 1) In the present study, whether or not the micro-lipophilicity of each ceramic could be influenced by the residence of microoil droplets has been addressed by making AFM observations, as follows. After a pre-selected surface area (3 mm 3 mm) of a specimen was observed in the ac non-contact mode [Fig. 6(a)], a larger oil droplet was attached to the tip of the cantilever of the AFM from elsewhere. Then, several ac non-contact mode scans were carried out in the same area (scanning speed: 4 nm/s) to disperse micro droplets of oil. The morphology of micro oil droplets and the specimen surface is shown in Fig. 6(b). As indicated by the arrow and the profile of the pore before and after oil droplet adhesion, it is clear that the micro-droplets can also reside in pores. Of course, larger numbers of micro-oil droplets stayed near the Fig. 6 Stay of oil droplets in pore ((a) before oil droplet dissemination (b) after oil droplet dissemination). pores. Hence, the micro-lipophilicity is influenced mostly by the surface composition rather than the geometric shape of the surface. Figure 7 shows the composition near the surface as analyzed by XRD. Besides the main component of Si 3 N 4,a trace of Si 2 N 2 O was detected in material, whereas no Si 2 N 2 O was detected in FM a material. Si 2 N 2 O was also detected in FM 1b and FM 1c, in which the addition of Fe 3 O 4 was greater than in FM a material. In EPMA analysis of, FM 1b and FM 1c materials, small particles in the micrometer range containing concentrations of Fe and Mo were uniformly distributed. In addition, areas of high oxygen concentration could be detected in other regions. It has been reported that Si 2 N 2 O can be produced by the oxygen from Fe 3 O 4 that was added to the Si 3 N 4 ceramics. 11) Figure 8 shows the oxygen concentration as a function of the amount of Fe 3 O 4 added to the FM and. It is clear that the oxygen concentration increased with increasing Fe 3 O 4. Furthermore, the oxygen concentration in was higher than that in FM a, even though their content of Fe 3 O 4 was the

5 Influence of Additives and Hot-Press Sintering on Mechanical and Lipophilic Properties of Silicon Nitride Ceramics 245 Intensity (arb. unit) Relative O Concentration (%) (Fe 3 O 4 :3.6mass%, Mo:.9mass%) (Fe 3 O 4 :3.6mass%,Mo:.9mass%) θ Fig. 7 XRD patterns of,, and. FM ) FM1b FM1c (FM HP ) Fe 3 O 4 (mass%) Fig. 8 Relationship between additional Fe 3 O 4 and O concentration in FM and by EPMA. same. This indicates that the oxygen concentration in the sintered ceramics can be influenced by hot-press sintering. Further, according to XPS analysis, Fe exists as Fe 3 O 4 and Mo exists as MoO 3. Since Si 2 N 2 O absorbs lubricating oil better than 11) Si 3 N 4, the formation of such a substance in Si 3 N 4 ceramics might improve both the macro- and microlipophilicity of the material. 4.3 Influence of Si 2 N 2 O on lipophilicity, friction coefficient and wear resistance As indicated above, we know that the Si 2 N 2 O component of,fm 1b and FM 1c materials might play an important role in their macro- and micro-lipophilicity. On the other hand, since the Fe 3 O 4 and MoO 3 particles are separated from each other by more than 5 mm on,fm 1b and FM 1c surfaces, 12) the micro-droplets of oil observed by AFM in a small area (5 mm 5 mm) would locate [preferentially] on areas containing Si 3 N 4, Fe 3 O 4, MoO 3 or Si 2 N 2 O. The relationship between each component and micro-lipophilicity can be summarized as follows: The small difference in micro-lipophilicity between and material indicates that Fe 3 O 4 and MoO 3 have little influence on micro-lipophilicity in FM a material. On the other hand, the improvement of micro-lipophilicity in material compared with FM a material, whose additions of Fe 3 O 4 and Mo are the same, can be attributed to the formation of Si 2 N 2 O in the material. Furthermore, the improvement of microlipophilicity in FM 1b and FM 1a can also be attributed to the formation of Si 2 N 2 O. Accordingly, macro-lipophilicity was mainly influenced by Fe 3 O 4, MoO 3 and Si 2 N 2 O, whereas micro-lipophilicity was mainly influenced by Si 2 N 2 O rather than Fe 3 O 4 and MoO 3. However, both the macro and micro contact angles of lubricating oil on were lower than those on, for reasons not entirely clear. The formation of Si 2 N 2 OinSi 3 N 4 ceramics might be also one of the reasons for the decrease in wear resistance, because the fracture toughness of Si 2 N 2 O is general lower than Si 3 N 4. However, its precise influence remains unclear. This shows that the addition of Fe 3 O 4 and Mo to Si 3 N 4 ceramics and hot-press sintering improve their lipophilic and mechanical properties (especially strength). However, the reinforcement of grain boundaries and smaller particle sizes are necessary for further improvement of strength. Clearly, the mechanism of formation of Si 2 N 2 OinSi 3 N 4 ceramics during hot-press sintering should receive further study. 5. Conclusions The addition of Fe 3 O 4 and Mo and the hot-press sintering process were used to prepare Si 3 N 4 ceramics to improve their lipophilic and mechanical properties, with the following results: (1) The bending strength, relative density, hardness and fracture toughness of Si 3 N 4 ceramics with added Fe 3 O 4 and Mo were improved by using the hot-press sintering process rather than the pressureless process. In particular, the bending strength and relative density were improved by about 2 and 14%, respectively. (2) Both the macro- and micro-lipophilicity of Fe 3 O 4 - and Mo-added ceramics were improved by using the hotpress process rather than the pressureless process. This can be attributed to the addition of Fe 3 O 4, and the formation of MoO 3 and Si 2 N 2 O during hot-press sintering. (3) Lower coefficient of friction and higher wear resistance of Fe 3 O 4 - and Mo-added ceramics have been achieved with the hot-press process, compared with ceramics of similar composition prepared by the pressureless process. The high wear resistance was considered to be due to the improvement of hardness and fracture toughness, which decreased the loss of particles during the wear test.

6 246 M. Kido, T. Tokuda, R. Wang and F. Suzumura REFERENCES 1) H. Kita, T. Iizuka, T. Hirai, T. Murao and Y. Unno: Ceramics 36 (21) ) T. Tokuda, K. Kato, R. Wang and M. Kido: J. Japan Inst. Metals 67 (23) ) T. Kinoshita: J. Japan Inst. Metals 59 (1995) ) M. Chu, S. Sato, Y. Kobayashi and K. Ando: J. Jpn. Soc. Mech. Eng. 61 (1995) ) R. Wang, M. Kido, T. Tokuda, K. Kato and S. Nakanishi: J. Japan Inst. Metals 67 (23) ) T. Kinoshita: J. Japan Inst. Metals 44 (1995) ) M. Iwasa and M. Kinoshita: Osaka Industry Technical Experiment Station, Japan 39 (1988) ) M. Iwasa and M. Kinoshita: Ceramics 24 (1989) ) T. Iizuka, T. Murao, H. Yamamoto and H. Kita: J. Ceram. Soc. Jpn. 19 (21) ) R. Wang, M. Takeda, K. Mukai and M. Kido: J. Japan Inst. Metals 65 (21) ) H. Kita, T. Iizuka, K. Osumi and T. Hirai: J. Ceram. Soc. Jpn. 111 (23) ) K. Kato, T. Tokuda, T. Dote and M. Kido: Proceeding of the 9th Young Ceramist Meeting in Chugoku and Shikoku (22)