Effects of HIP Treatment on the Microstructure and Properties of Cr35-Si65 Target

Transcription

1 Materials Transactions, Vol. 50, No. 2 (2009) pp. 395 to 400 #2009 The Japan Institute of Metals EXPRESS REGULAR ARTICLE Effects of HIP Treatment on the Microstructure and Properties of Cr35-Si65 Target Chung-Hung Tam 1; *, Shih-Chin Lee 1, Shih-Hsien Chang 2 and Fong-Cheng Tai 3 1 Department of Materials Science and Engineering, National Cheng Kung University, Tainan, 701, Taiwan, R. O. China 2 Department of Materials and Mineral Resources Engineering, National Taipei University of Technology, Taipei, 106, Taiwan, R. O. China 3 Sustainable Environment Research Center, National Cheng Kung University, Tainan, 709, Taiwan, R. O. China Hot Isostatic Pressing (HIP) can be used for upgrading castings, densifying pre-sintered components, consolidating powders, and interfacial bonding. Hot pressing is a good method for fabricating on high melting materials with better mechanical properties. Commercial ashp treated Cr35-Si65 targets are used throughout this study. The aim of this paper is to discuss the methods and to find a suitable HIP treatment for the as-hp treated Cr35-Si65 Meanwhile, we also to evaluate the effects of HIP treatment on the microstructure and properties of Cr35- Si65 The experiment results show that HIP treatment at 1100 C under 175 MPa for 2 hours of Cr35-Si65 target is the most suitable condition. It can reduce the open pore porosity by 20% and close pore porosity by 30%. The density could be increased to 3: kg/m 3. [doi: /matertrans.mer ] (Received August 8, 2008; Accepted November 25, 2008; Published January 25, 2009) Keywords: hot isostatic pressing, hot pressing, Cr35-Si65 target, porosity, density 1. Introduction Hot pressing (HP) is putting the powder in a die and applying force with simply uniaxial pressure. The hot press method is convenient to fabricate an alloy target because it is advantageous to form an arbitrary shape and to choose many additive elements. Dense and fine grain material can be obtained by hot pressing through rapid heating of the powder materials under moderate loads. 1 3) However, this method generally brings some oxygen and nitrogen into a target and is not adequate for obtaining a commercial Cr-Si 3,4) Chromium silicides are mainly applied in thin film resistor and heater. Since Cr and Si are active and brittle, when the target is produced by casting, both elements would be oxidized and the target may have crack during casting. The poor quality such as purity and density of the targets are unacceptable for thin film deposition. Thus we need to make the Cr-Si targets by other advanced methods like hot isostatic pressing. The products made by hot pressing may still be porous, and the size is limited by the die. For denser and larger targets, we can apply hot isostatic pressing to achieve our aim. 5 8) HIP combination of pressure and temperature can be used to achieve a particular density at a lower temperature than would be required for sintering alone and at a lower pressure than required for cipping. 4) The effect of the lower temperature is that unacceptable grain growth can be avoided. In addition, the methods identified previously for enhancing densification of powders by introducing additives such as low-melting-point constituents which may have deleterious effects on mechanical properties are not needed. 9) It is a material pressing technique that high isostatic pressure is applied to a powder part or compact at elevated temperature to produce particle bonding. 10,11) Functions of HIP processes include defect healing of castings, removal of internal porosity, diffusion bonding of *Corresponding author, asbeltam@so-net.net.tw dissimilar materials, improvement in property scatter and mechanical properties ) In this study, we were afraid that the Cr-Si target would react with nitrogen or oxygen during HIP, thus we used argon as the atmosphere. We took three different temperatures and two kinds of soaking time for HIP treatment. 2. Experimental Procedures Both of chromium and silicon are rapidly oxidized, and powder metallurgy is usually used to make targets out of such materials, especially in hot pressing. To achieve better quality targets, we should use some advanced methods such as hot isostatic pressing. HIP consolidates powders of metals, ceramics, and carbides into fully dense, complex parts with net or near net shapes. 15,16) In earlier studies, we had reported that the HIP treatment could eliminate the open and close pores of targets. Cr50- Si50 target decreased to 40% in porosity after 1100 C, 175 MPa, 4 hours HIP treatment. 17) The aim of this paper is continuous to discuss the effects of HIP treatment on microstructure and properties for Cr35-Si65 Commercial as-hp treated Cr35-Si65 (Cr-65 mass%si) targets were used throughout this study, they were made by a 1000 C, 50 MPa die. After hot pressing process, we would put the targets into HIP furnace to enhance the microstructure and properties. Meanwhile, all of the targets were capsulefree. Both of the crucibles and HIP heaters are made by graphite. Avoiding the solid-to-solid contact of crucibles and targets, we applied alumina plates to put on the bottom of targets. Before running the HIP process, we draw air away from the vessel to 1: torr, then to get the cleaner operating environment. The HIP schedule was shown, as Fig. 1. The pressing and heating were applied concurrently. The final temperatures were 1050 C, 1100 C and 1150 C, and the final pressure was 175 MPa, and the soaking times were 2 and 4 hours. To evaluate the effects of microstructure and properties of the Cr35-Si65 target by HIP process, SEM,

2 396 C.-H. Tam, S.-C. Lee, S.-H. Chang and F.-C. Tai Fig. 1 HIP processing schedule. Fig. 2 Comparison of the porosities of as-hp treated and after different HIP treatment for Cr35-Si65 XRD, TEM, porosity and density inspections were performed. SEM images are formed through secondary and backscattered electrons as secondary electron image (SEI) and back-scattered electron image (BEI). The Cr and Si contents of the phases were detected by microanalysis via EDS, and the results would compare with SEI and BEI. The open pore porosity and appearance density was measured by ASTM C standard. 18) Moreover we measured the real density by pycnometer; the specimen was crashed into powder by agate mortar and passing the 325 screen mesh. We calculated the closed pore porosity by comparing the open porosity, appearance density and pycnometer density (real density). Before we filled argon as the HIP atmosphere, there was little residual air about nitrogen and oxygen in the vessel. The residual air might store in the porosity or form some compounds as the impurities inside the Cr-Si targets. They would be measured by the concentrations of nitrogen and oxygen analyzer with vaporizing the small chips from them. 3. Results and Discussion The porosity test followed ASTM C standard and applied specific gravity bottles. During HIP plastic flow, power law creep and grain boundary sliding in the particles and bulk diffusion in the particle contacts may contribute to the densification. 5) Figure 2 represents the porosity of Cr35- Si65 target of as-hp treated and after HIP treatment at different temperatures and times. It shows the porosity of open pores was reduced at 1050 C and 1100 C, 2 hours HIP treatment, but the porosity would rise back after the 4 hours HIP treatment, both of 1050 C and 1100 C. The porosity of close pores would reduce after HIP treatment. When the soaking time is extended from 2 hours to 4 hours at 1050 Cor 1100 C, the close pore porosity would decrease a little more. The open pores would be filled and covered with the diffusive materials by HIP, and some of them would become close ones. The materials diffused into the close pores because of high temperature and pressure. When the soaking time is longer, the diffusion becomes more and more and eliminates the close pores completely. On the other hand, the main materials of Cr35-Si65 are CrSi 2, which is reacted from Cr and Si, and Si. Silicon is easy to adsorb the gas at high Fig. 3 Comparison of the pore size of as-hp treated and after 1100 C, 175 MPa, 2 h. HIP treatment for Cr35-Si65 temperature. Thus the open pore porosity would not reduce efficiently. Silicon would form larger open pores or new ones through gas adsorption. With the HIP time extended, the open pore porosity would even rise back and higher than the as-hp one. The porosity of open pore for Cr35-Si65 target was reduced down to 80% from the as-hp one after HIP treatment at 1100 C, 2 hours. And the close pore porosity was reduced down to 70%. When the temperature of HIP was increased to 1150 C, the open pore porosity rapidly increased. This is because the silicon of Cr35-Si65 target was overheated in the high temperature and pressure environment, and the surface materials melted partially. There were more and more chances for silicon to adsorb the gas. And the gas can get into the target easier at 1150 C than the temperature at 1050 C or 1100 C. After HIP, the target was cooling down, and the gas would form more open pores. Figure 3 shows the pore size distribution detected by mercury intrusion porosimetry (SCIENTEK CORPORATION, STK019348) of as-hp treated and after HIP treatment at 1100 C with 2 hours. For the as-hp Cr35-Si65 target, the main pore sizes were distributed at about 700 nm and 10 nm. The 10 nm pores would be significantly eliminated after HIP treatment. But the 700 nm ones would become a little

3 Effects of HIP Treatment on the Microstructure and Properties of Cr35-Si65 Target 397 Fig. 4 Comparison of the density of as-hp treated and after different HIP treatment for Cr35-Si65 Fig. 5 Comparison of the nitrogen and oxygen concentration of as-hp treated and after HIP treatment at different temperatures for Cr35-Si65 more and larger since the gas adsorption of silicon. The porosity is smoothed away and reduced after HIP treatment totally. Figure 4 presents the Cr35-Si65 target density of as-hp treated and HIP treatment at different temperatures and times. It shows that the density increases after HIP. The longer soaking time we dealt gets the lower density at the same HIP temperature, since silicon would adsorbs more gas to form more and large pores. At the temperature of 1100 C and the soaking time of 2 hours, we can get the optimal density about 3: kg/m 3. In general the density increases with the HIP temperature climbing up, but it became lower for the temperature raised from 1100 C to 1150 C because of the significantly increasing amount of open pores by silicon adsorption. The results are calculated by the ASTM C standard after the porosity testing. Thus the density is related to the porosity showed in Fig. 2. Figures 5 shows the nitrogen and oxygen concentrations of as-hp treated and after HIP treatment at different temperatures and times for Cr35-Si65 The vacuum before filling argon as the HIP atmosphere and heating up is about 10 3 torr. There would be still little residual gas about oxygen and nitrogen inside the vessel and targets. It is hard for the system to draw all the nitrogen and oxygen away before running the HIP process. When the HIP process started, the gas stored inside the targets would be released by heating. K. Ishizaki reported that the oxygen concentration would remain almost constant with the increasing HIPing temperature. 19) Moreover, due to the heaters and crucibles were made by graphite. According to the Ellingham Diagram, 20,21) carbon oxides can exist at any equilibrium oxygen partial pressure with different ratio of CO/CO 2.If there is enough carbon in the experimental atmosphere, the partial pressure of oxygen is governed by carbon. 20,22,23) Graphite heater, this kind of carbon would not release almost any gases, but they would govern the oxygen partial pressure to oxidize Cr or Si. Therefore it would cause the oxygen concentration increased. Although we applied the alumina plates to avoid the solid-to-solid contact between crucibles and targets, the gas such as oxygen would still contact and react through the HIP atmosphere. Even the partial pressure of oxygen is very low. The experimental results show that the concentrations of oxygen impurities would be almost constant but slightly increased after HIP treatment. When the soaking time became longer, the oxygen concentration would also become a little higher. Therefore the amount was not enough to change the pycnometer density of Cr35-Si65 The nitrogen concentrations would be almost constant without changing on the Cr35-Si65 target after HIP treatment. Silicon exists in the Cr35-Si65 target, and it would adsorb the gas during HIP process. Silicon might even react with little residual oxygen and nitrogen to form oxides and nitrides. The results show that oxygen is more active than nitrogen with silicon. Nitrogen and oxygen may still store in the closed pores of the target after HIP, and they should be physical adsorbed or become some nitrides and oxides, even they are too little to change the pycnometer density. The amount of nitrides and oxides is not much enough that we cannot find them out with XRD as long-range order crystals. Moreover, nitrides and oxides can be thin films. X-ray does not prove that they did not exist. We may purge the vessel with argon with several cycles to reduce the residual air or use some dies like canning to protect the target from some reactive gas during HIP. Thus we can get the purer and denser Cr-Si Figure 6 represents the XRD patterns of as-hp treated and HIP treated at different temperatures Cr35-Si65 According to the Cr-Si phase diagram, 24) the composition of Cr35-Si65 (Cr-65 mass%si) forms CrSi 2 and Si, and the patterns show the same as the phase diagram. The patterns have little change after HIP. It means that the crystal structures have been stable after hot pressing. In this case, HIP after hot pressing improves the target density physically. It shows that CrSi 2 is stable after it became silicides from Cr and Si powder with hot pressing, and silicon would not change its crystal structure by gas adsorption. It shows that HIP would improve the density and porosity, but not change the composition and crystal structure to influence the thin film deposited by the Cr-Si Figure 7 are the secondary electron images (SEI) of as-hp treated and HIP treated at different temperatures Cr35-Si65

4 398 C.-H. Tam, S.-C. Lee, S.-H. Chang and F.-C. Tai Fig. 6 XRD patterns of as-hp treated and after HIP treatment (175 MPa, 4 h.) at different temperatures for Cr35-Si65 target (a) as-hp treated, (b) 1050 C, (c) 1100 C, (d) 1150 C. targets. It is easily observed that there are many large spherical closed pores inside the target as hot pressing by Fig. 7(a). Increasing HIP temperature can reduce the original large pores; both of the HIP temperature at 1050 C as Fig. 7(b) and 1100 C as Fig. 7(c) show that the original large spherical closed pores are eliminated successfully. When the HIP temperature increased to 1150 C as Fig. 7(d), the surface materials, especially silicon would become more active and diffusive due to the higher temperature. The HIP atmosphere were involved and trapped by silicon. It brought more small open pores generating and the porosity increased. Figure 8 show the SEI and backscattered electron images (BEI) of as-hp treated and 1100 C, 175 MPa, 2 hours HIP treatment for the Cr35-Si65 From the EDS results, the BEI images as Fig. 8(b) and 8(d) represent that the gray area is CrSi 2 and charcoal gray area is Si. In Fig. 8(b), it presents that most of the pores exist inside the silicon part of Cr35- Si65 target after hot pressing. After HIP treatment as Fig. 8(d), we observe that the pores inside the silicon part had become smaller and fewer. It shows HIP treatment can enhance the density and reduce the porosity of the Cr35-Si65 Figure 9 shows HRTEM micrographs and pattern results of CrSi 2 [101] zone of Cr35-Si65 target after 1100 C, 175 MPa, 2 hours HIP treatment. The CrSi 2 crystal structure is hexagonal determined from the diffraction pattern. It also can observe that CrSi 2 is lamellar structure existed in the Cr35-Si65 target; the CrSi 2 crystal is near perfect after HIP treatment. This is good for application such as thin film deposition. 4. Conclusion (1) HIP treatment is helpful for eliminating the open and close pores of targets made by hot pressing. Cr35-Si65 target decreased to 80% in open pore porosity and 70% in close pore porosity after 1100 C, 175 MPa, 2 hours HIP treatment. The density can be improved to 3: kg/m 3. (a) (b) (c) (d) Fig. 7 SEM micrographs of as-hp treated and HIP treatment (175 MPa, 4 h.) at different temperatures for Cr35-Si65 target (a) as-hp treated, (b) 1050 C, (c) 1100 C, (d) 1150 C.

5 Effects of HIP Treatment on the Microstructure and Properties of Cr35-Si65 Target 399 (a) (b) CrSi 2 Si (c) (d) CrSi 2 Si Fig. 8 Comparison of the SEI and BEI micrographs of (a)(b) as-hp treated and (c)(d) 1100 C, 175 MPa, 2 h. HIP treatments for Cr35-Si65 Fig. 9 HRTEM micrograph and associated electron diffraction of CrSi 2 [101] for Cr35-Si65 target after 1100 C, 175 MPa, 2 h. HIP treatment. (2) In the research, due to the partial pressure of oxygen is governed by carbon to oxidize the targets; oxygen concentrations of the Cr35-Si65 target would be almost constant but slightly increased after HIP treatment. However, nitrogen concentrations would be almost constant without changing. Acknowledgments This research supported by the Materials and Electro- Optics Research Division of Chung-Shan Institute of Science and Technology Taiwan. The authors would also like to express their appreciation to ASSAB STEELS TAIWAN CO., LTD.

6 400 C.-H. Tam, S.-C. Lee, S.-H. Chang and F.-C. Tai REFERENCES 1) W. Arbiter: WADC technical report part I (1953) ) W. Y. Kim, H. Tanaka, A. Kasama, R. Tanaka and S. Hanada: Intermetallics 9 (2001) ) J. Meng, C. Jia and Q. He: J. Alloy. Comp. 421 (2006) ) A. Kikitsu, T. Maeda, T. Kai, T. Nagase, J. Akiyama, N. Fujioka and T. Ishigami: IEEE Trans. on Magnet. 39 (2003) ) Yu. N. Taran-Zhovnir: Mat. Sci. Heat Treat. 40 (1998) ) J. C. Kuang, C. R. Zhang, X. G. Zhou and S. Q. Wang: Mater. Res. Bull. 39 (2004) ) M. Naka, M. Maeda, T. Shibayanagi, H. Yuan and H. Mori: Vacuum 65 (2002) ) P. Wei and X. R. Wan: Surf. Coat. Technol. 132 (2000) ) H. V. Atkinson and S. Davie: Metall. Mater. Trans. A 31A (2000) ) S. Haro, R. Colas, Abraham Velasco and David Lopez: Mater. Chem. and Phys. 77 (2003) ) E. Klar: Metals Handbook Ninth Edition Vol. 7 Powder Metallurgy, (American Society for Metals, 1984) pp ) T. Garvare: Proceedings of the International Conference on HIP-Lulea/ June, (The Swedish Institute of Production Engineering Research, 1987) pp ) M. Koizumi: Hot Isostatic Pressing Theory and Applications, International Conference on Hot Isostatic Pressing, (Elsevier Applied Science, 1992) pp ) Flow Autoclave System, Inc.: Hot Isostatic Pressing Applications (1996). 15) H. Bei, E. P. George and G. M. Pharr: Intermetallics 11 (2003) ) S. C. Lee, S. H. Chang, T. P. Tang, H. H. Ho and J. K. Chen: Mater. Trans. 47 (2006) ) C. H. Tam, S. C. Lee, S. H. Chang, T. P. Tang, H. H. Ho and H. Y. Bor: Mater. Trans. 49 (2008) ) ASTM C (2006). 19) K. Ishizaki: Hot Isostatic Pressing Theory and Applications, International Conference on Hot Isostatic Pressing, ed. by M. Koizumi, (Elsevier Applied Science, 1992), pp ) K. Ishizaki: Acta. Met. Et Mat. 38 (1990) ) H. Li and G. S. Selvaduray: Ellingham Diagram Web Project, Dept. of Chem. & Mat. Eng., San Jose State University, sjsu.edu/ellingham/ 22) K. Ishizaki: Hot Isostatic Pressing Theory and Applications, International Conference on Hot Isostatic Pressing, ed. by R. J. Schafer and M. Linzer, (ASM International, Metals Park, Ohio, 1991), pp ) K. Ishizaki: Euro-Ceramics, Processing of Ceramics, Vol. 1 Ed. by G. de With, R. A. Terpstra and R. Metselaar, (Elsevier Appl. Sci., London and New York, 1989) pp ) T. B. Massalski (ed): Binary Alloy Phase Diagrams, 2nd Edition, Vol. 3 (Materials Park, OH: ASM International, 1990).