Effect of Consolidation Process on Tensile Properties of Fe Cu P/M Alloy from Rapidly Solidified Powder

Transcription

1 Materials Transactions, Vol. 44, No. 7 (2003) pp to 1315 Special Issue on Growth of Ecomaterials as a Key to Eco-Society #2003 The Japan Institute of Metals Effect of Consolidation Process on Tensile Properties of Fe Cu P/M Alloy from Rapidly Solidified Powder Hideki Kakisawa*, Kazumi Minagawa, Susumu Takamori, Yoshiaki Osawa and Kohmei Halada National Institute for Materials Science, Tsukuba , Japan The possibilities for recycling scrap iron utilizing copper impurity as a reinforcement were investigated. For this purpose, Fe Cu powder was prepared by high-pressure water atomization and consolidated by groove rolling at a warm temperature in order to maintain the powder microstructure. The effect of the rolling reduction rate on the mechanical properties of the consolidated samples was examined in this paper. The powder was successfully consolidated without coarse copper precipitation by the rolling technique, though pores were observed at the primary powder boundaries at the low reduction rates. With the increase of the reduction rate, the powder was more elongated and adhered with other powder, and the pore volume fraction decreased. Strength and elongation in tensile testing were measured, and the Vickers hardness of the consolidated samples was examined. All the properties tested improved by increasing the reduction rate. The tensile properties required a more intensive rolling process to bring about the full performance originated by the maintained powder microstructure than the hardness. Using the determined optimum rolling condition, a high strength material was achieved via the proposed processing, showing the prospect for a new recycling process of scrap iron. (Received January 27, 2003; Accepted March 27, 2003) Keywords: Iron copper rapidly solidified powder, rolling consolidation, reduction rate, tensile properties 1. Introduction *Corresponding author: KAKISAWA.Hideki@nims.go.jp Copper, a major impurity in scrap iron from vehicles and electrical appliances, is one of the most troublesome elements to recycling. It is very difficult to remove, by refining, copper that is contained in iron. The liquid copper phase in iron causes surface cracking during hot rolling; it is known that the limitation of copper in rolled bar steel is 0.4 mass%. Copper in the amount of 0.4 mass% in rolled bar steel will cause surface cracking to occur. 1) The embrittlement of steel products due to coarse copper precipitation is also a problem. Iron scrap with significant copper impurity has been recycled in casting at best, or has been dumped without recycling. However, a huge amount of scrap iron is being produced, and the copper content in iron scrap has continued to increase over the years. 2) It is necessary to develop a new type of processing so that copper rich scrap can be recycled without removing the copper. New uses for copper-rich scrap iron, in addition to casting, are also needed. We have been investigating the utilization of copper impurity as a reinforcement using a powder metallurgical technique. In powder metallurgy, an Fe Cu system has often been used. Liquid phase sintering is used for the consolidation of iron powder, utilizing melted copper surrounding iron powder. 3 5) Fe Cu powders in which copper has been supersaturated by mechanical alloying are consolidated in various methods at relatively low temperatures of 873 K. 6 9) The immiscibility of these metals produces an ultrafine microstructure, and the powder microstructure which is maintained due to the low consolidating temperature results in nanostructured materials with good mechanical properties. However, much time and energy is required to prepare the raw powder used in processes that use mechanical alloying. In our study, a rapidly solidified Fe Cu powder from high-pressure water atomization, where copper is nanodispersed, was used ) It has been known that precipitation hardening operates when copper precipitates in ferrite matrix on the order of nanometers ) By consolidating at 873 K using the rolling technique, the copper impurities remained nano-dispersed in the iron matrix throughout the process without either liquid phase or coarse precipitation, and the consolidated materials showed high strength. 10,11) In this paper, the effect of the reduction rate on the consolidation procedure of the powder was investigated, and the optimum consolidation condition was determined from the results obtained from tensile testing. 2. Experimental Procedure Fe Cu alloy powder (Nippon Atomized Metal Powders Corporation, Tokyo, Japan) was prepared using a highpressure water atomizing method with a copper content, c cu, of 0.5, 2 and 5 mass%. The atomized powder was sorted, and the mean powder size was set to 5 mm. In X-ray diffraction analysis of the as-atomized powders, peak sets of fcc copper and ferrite were detected, indicating that the copper was precipitated in the ferrite matrix. Figure 1 shows the cross section of the powder (Cu: 0.5 mass%) observed by scanning electron microscopy (SEM). There were no coarse copper precipitates either in the powder or on the powder surface; most of the precipitated coppers were less than 20 nm in diameter. The powder with 2 mass% copper was used to determine the optimum consolidation condition. The powder was filled in a low carbon steel sheath, and was sealed after being evacuated at 753 K for 54 ks in order to remove the surface contaminant of the powder. The sheath had an outer diameter of 40 mm and an inner diameter of 30 mm. The volumetric capacity of the sheath was 84.3 cm 3, and approximately 330 g of the powder were put into the sheath. The sheath was then heavily deformed and consolidated with groove rolling at 873 K. The 40 mm diameter sheaths were formed into bars,

2 1312 H. Kakisawa, K. Minagawa, S. Takamori, Y. Osawa and K. Halada Fig. 1 SEM observation of the cross-section of the atomized powder (Cu: 0.5 mass%). Table 1 Rolling path and cross-section reduction rate. Rolling path Bar diameter (mm) cross-section reduction rate and the powder was consolidated through the rolling process. The rolling paths are listed in Table 1. During the rolling, the sheath were re-heated at 873 K for 300 s every three steps to avoid a temperature drop. The 3-step rolling took less than 60 s. The cross-section reduction rate, R, defined by R ¼ S 0 S 100ð%Þ ð1þ S 0 where S 0 is the initial cross-section of the bar, and S is the cross-section after rolling, was varied, and bars with R ¼ 65:4, 71.6, 76.7, 80.8, and 87.2 were obtained. After rolling, the consolidated samples were cooled in air. Transverse and longitudinal sections of the consolidated samples were observed by scanning electron microscopy (SEM). Electron probe microanalysis (EPMA) of the sections was conducted. Vickers microhardness was measured with an applied load of 500 N and a holding time of 15 s. Tensile testing was done in air at room temperature using round tensile test specimens that had a parallel span of 24.5 mm and a diameter of 3.5 mm. The crosshead speed was a constant rate of 8: mm/s. Note that the primary outer sheath parts were completely removed when the bars were machined to specimens with a diameter of 3.5 mm. After the testing, SEM observations of the fracture surface and longitudinal section of the fractured specimens were carried out. The powder with 0.5 and 5 mass% copper was also consolidated in the same conditions, with the cross-section reduction rate fixed on R ¼ 87:2. The consolidated samples were subjected to tensile testing under the same test conditions mentioned above. Fig. 2 SEM observation of the longitudinal sections of the consolidated samples (Cu: 2 mass%): (a) R ¼ 65:4, (b) R ¼ 76:7, and (c) R ¼ 84:4. 3. Results and Discussion Figure 2 shows the longitudinal sections of the consolidated samples (Cu: 2 mass%): (a) R ¼ 65:4, (b) R ¼ 76:7, and (c) R ¼ 84:4. Consolidation was successfully completed without crumbling in all the ranges of R tested. The copper nano-dispersed microstructure of the powder was maintained after consolidation in all of the samples. The primary powder boundaries remained, and pores were observed at the boundaries. Some of the powders had an oxide layer around

3 Effect of Consolidation Process on Tensile Properties of Fe Cu P/M Alloy from Rapidly Solidified Powder 1313 them, which formed during water atomization (indicated by arrows). Applying the concentration histogram imaging (CHI) method 20) to the EPMA data, most of the inclusions were found to be Fe 2 O 3. With the proceeding of the rolling process, the powders were severely deformed and adhered with each other, and the pores at the boundaries decreased. Some of the oxide layers were fractured by the intensive deformation. Figure 3 shows the relation between the pore volume fraction, V p, and the cross-section reduction rate, R. In the sample rolled at R ¼ 87:2, a relative density of 99.8% was obtained. Since the powder packing density in the sheath before consolidation was about 50%, assuming the density of -iron at room temperature 7.8 g/cm 3, the data indicate that good consolidation was achieved though the rolling used in this study. Figure 4 shows the relation between the Vickers microhardness, H v, in the transverse section, and the cross-section reduction rate, R. The hardness increased with the increase of the cross-section reduction rate, i.e., the proceeding of consolidation. The hardness reached 240 at R ¼ 71:6 and then became constant. The high hardness of 240 was attributed to the hard microstructure of the powder where Fig. 3 Relation between pore volume fraction in the consolidated samples and cross-section reduction rate. Fig. 4 Relation between Vickers microhardness and cross-section reduction rate. Fig. 5 Ultimate tensile strength and elongation of the consolidated samples plotted against cross-section reduction rate. copper was precipitated, and the constant hardness in R 71:6 indicated that the cross-section reduction rate of 71.6 was enough to utilize the powder microstructure for hardening. The relation between ultimate tensile strength, u, and the cross-section reduction rate, R, is shown in Fig. 5. The ultimate tensile strength increased as the rolling proceeded. The sample rolled at R ¼ 87:2 had a high strength of 753 MPa while the strength of the sample rolled at R ¼ 71:6 was 688 MPa, though they had the same hardness. The relation between uniform elongation in tensile testing, ", and R is also plotted in Fig. 5. The elongation had significant dependence on the reduction rate. The samples rolled at R ¼ 65:4 and 71.6 fractured on reaching the maximum stress, resulting in a small elongation. The samples rolled at the higher rate had a larger uniform elongation, and after the maximum stress, necking behavior occurred. Both the uniform and total elongation increased with the increase of the reduction rate in this reduction rate range. The longitudinal section of the samples after testing is shown in Fig. 6: (a) R ¼ 65:4 and (b) R ¼ 84:4. The end of the sample is the fracture part. This figure clearly shows that the fracture occurred at the primary powder boundaries in the samples rolled at the low cross-section reduction rates. Weak adhesion with the neighboring powders due to a large number of pores and the oxides caused the fracture at the boundaries, i.e., the samples fractured before exhibiting the full potential expected from the copper nano-dispersed microstructure of the powder. The small elongation is also explained by the boundary fracture. In the samples rolled at R ¼ 87:2%, each powder was well deformed until it reached fracture without boundary breaking. Though small pores of 500 nm grew at the boundaries, these did not lead to the boundary breaking directly. This fracture behavior is attributable to two main reasons, both of which come from the larger deformation of the powders during rolling. One reason is the smaller boundary section concerning the fracture, i.e., the boundary section perpendicular to the loading direction, and the other reason is the breakage of the oxide layer and the creation of fresh surfaces that lead to stronger boundary adhesion, schematically illustrated in Figs. 7(a) and (b). The surface

4 1314 H. Kakisawa, K. Minagawa, S. Takamori, Y. Osawa and K. Halada Fig. 6 SEM observation of the longitudinal sections of the fractured samples: (a) R ¼ 65:4 and (b) R ¼ 84:4. area of the deformed powder, A, was calculated by assuming that the volume of the powder was constant and the sphere powder was deformed to an ellipsoid form, as shown in Fig. 7(c). The surface area increases as the major axis, a, is extended to the rolling direction. From all of these results, it was determined that severer rolling is required to bring out the inherent high strength of the samples than is required to obtain the full hardness, and the optimum rolling condition was determined to be R ¼ 87:2. The ultimate tensile strength and elongation of the samples consolidated in the optimum condition are plotted against the copper content in Fig. 8. The strength increased with the increase of the copper content in the samples, clearly indicating the contribution of copper to strengthening. The elongation decreased as the copper content increased. The sample with 5 mass% had a significantly high strength of 867 MPa. The samples had considerable elongation in all the ranges of copper content: 11.7% total elongation and 3.9% uniform elongation at 0.5 mass% copper; 6.9% total elongation and 2.7% uniform elongation even at 5 mass% copper, exhibiting a very good strength/elongation balance. A high strength material utilizing a copper additive for strengthening has been achieved by the combination of rapidly solidified Fig. 7 Schematic illustrations of boundary in the consolidated samples, showing the smaller boundaries concerning the fracture and the larger fresh surfaces in the samples rolled at a high reduction rate. Fig. 8 Relation between ultimate tensile strength and elongation of the consolidated samples rolled at R ¼ 84:4 and copper content.

5 Effect of Consolidation Process on Tensile Properties of Fe Cu P/M Alloy from Rapidly Solidified Powder 1315 powder and a severe rolling consolidation technique, and the process is applicable up to a copper content of 5 mass%, which is an extremely high amount as a tramp element. This indicates the great possibilities of this method for supplying a new use for scrap iron which includes significant copper. 4. Conclusion (1) Consolidation of the powder proceeds as the reduction rate increases. A higher reduction rate results in a larger powder deformation and smaller pore volume fraction. (2) The larger deformation led to the generation of fresh surface on the powder and resulted in strong adhesion at the boundaries. The strong adhesion of each powder was the key to bringing out the full performance originated by the maintained copper nano-dispersed powder microstructure. More rolling was required to achieve the best tensile properties than hardness. (3) The prospect for supplying a new use for scrap iron which includes significant copper, as has been applied only to casting, was clearly demonstrated. REFERENCES 1) N. Imai, N. Komatsubara and K. Kunishige: ISIJ Int. 37 (1997) ) K. Noro, M. Takeuchi and Y. Mizukami: ISIJ Int. 37 (1997) ) W. D. Kingery and M. D. Narasimhan: J. Appl. Phys. 30 (1959) ) W. A. Kaysser, W. J. Huppmann and G. Petzow: Powder Metall. 23 (1980) ) S. S. Kang and D. N. Yoon: Metall. Trans. 13A (1982) ) G. R. Shaik and W. W. Milligan: Metall. Mater. Trans. 28A (1997) ) J. E. Carsley, A. Fisher, W. W. Milligan and E. C. Aifantis: Metall. Mater. Trans. 29A (1998) ) L. He and E. Ma: Nanostruct. Mater. 7 (1996) ) J. E. Carsley, W. W. Milligan, S. A. Hackney and E. C. Aifantis: Metall. Mater. Trans. 26A (1995) ) E. Hornbogen and R. C. Glemn: Trans. Metall. Soc. AIME 218 (1960) ) K. G. Russel and L. M. Brown: Acta Metall. 20 (1972) ) S. R. Goodman, S. S. Brenner and J. R. Low, Jr.: Metall. Trans. 4 (1973) ) K. Minagawa, H. Kakisawa, H. Okuyama, M. Otaguchi and K. Halada: Proc Powder Metallurgy World Congress, ed. by K. Kosuge and H. Nagai (Jpn. Soc. Powder and Powder Metallurgy, Kyoto, Japan, 2000) pp ) H. Kakisawa, K. Minagawa, M. Otaguchi and K. Halada: Mater. Trans. 43 (2002) ) H. Kakisawa, K. Minagawa and K. Halada: Mater. Sci. Eng. A340 (2003) ) G. M. Worrall, J. T. Buswell, C. A. English, M. G. Hetherington and G. D. W. Smith: J. Nuclear Mater. 48 (1987) ) K. Osamura, H. Okuda, S. Ochiai, M. Takashima, K. Asano, M. Furusawa, K. Kishida and F. Kurosawa: Iron Steel Inst. Jpn. Int. 34 (1994) ) Y. Tomita, T. Haze, N. Saito, T. Tsuzaki, Y. Tokunaga and K. Okamoto: ISIJ Int. 34 (1994) ) A. Takahashi and M. Iino: ISIJ Int. 36 (1996) ) D. S. Bright and D. E. Newbury: Anal. Chem. 63 (1991) 243A-250A.