Semisolid Extrusion of Low-Carbon Steel

Transcription

1 Materials Transactions, Vol. 48, No. 4 (2007) pp. 807 to 812 #2007 The Japan Society for Technology of Plasticity Semisolid Extrusion of Low-Carbon Steel Sumio Sugiyama 1, Jingyuan Li* and Jun Yanagimoto 1 1 Institute of Industrial Science, The University of Tokyo, Tokyo , Japan The semisolid state behavior and semisolid extrusion properties of low-carbon steel were investigated, focusing on the possibility of clarifying a semisolid-forming process. First, the cooling curve of low-carbon steel is assessed by thermal analysis, so as to clarify the semisolid temperature range. The microstructure of the hot-rolled bar obtained at the semisolid temperature is the globular structure similar to those of aluminum and magnesium alloys. Then, to obtain a better understanding of the semisolid deformation behavior of the material, extrusion tests are carried out at various billet temperatures and cooling conditions at the die exit. To prevent the temperature decrease of the billet, a graphite case and block surrounding the billet acted as an insulator in the extrusion tests. The mechanical properties and the microstructure of the extruded products are evaluated and discussed, i.e., the extrusion force in the semisolid state is less than half that of the hot extrusion. The distribution of chemical components, such as carbon, is measured in the radial direction of the cross section of the products. The roomtemperature hardness of the as-extruded products shows a specific distribution from the surface to the center. The hardness at the center is approximately two times greater than that at the surface. [doi: /matertrans ] (Received January 20, 2006; Accepted December 10, 2006; Published March 25, 2007) Keywords: semisolid, extrusion, low-carbon steel, microstructure change, gradient composite 1. Introduction From the standpoint of saving natural resources and production costs, it is very important to optimize metalforming processes; semisolid metal forming has played a very important role in this aspect. Recently, aluminum and magnesium alloys have been directly produced from moltenmetal and semisolid-metal forming to eliminate conventional processes such as hot forming. 1) However, research on the semisolid metal-forming of iron alloys is still at an early stage. Research on the semisolid forming of iron alloys started at the beginning of the 90s. In semisolid forging processes, Kapranos et al. 2) studied the semisolid forging of tool steels, Midson et al. 3) introduced stainless-steel components produced by semisolid metals (SSM) forming strain-induced melt-activated (SIMA) processed alloys, Kiuchi et al. 4) reported on the semisolid forging properties of cast iron, and Nohn et al. 5) and Kopp et al. 6) discussed the thixoforming of steels (C70S6 and HS6-5-2). In semisolid rolling processes, Kiuchi and Sugiyama 7) reported the semisolid rolling of cast irons, Yoshida et al. 8) clarified the advantages of strip rheocasting (SRC) stainless steel, and Kang et al., 9) studied the direct rolling of semisolid-steel (0.6 (mass%)c and 1Cr18Ni9Ti) slurry. In semisolid extrusion processes, Secordel et al. 10) carried out experimental extrusion tests to study the rheological behavior of semisolid steels (100C6 and Z85WDCV6542), Miwa and Kawamura 11) published information on the semisolid extrusion forming of stainless steel, Abdelfattah et al. 12) and Rouff et al. 13) extruded 0.8 (mass%)c steel-based material in a semisolid temperature range, and Robelet et al. 14) reported direct extrusion test results of C38 and C80 steels. There have been several semisolid-forming studies using cast iron, tool steel and stainless steel, whereas no research on *Graduate Student, The University of Tokyo. Present address: Faculty of Materials Science and Engineering, University of Science and Technology Beijing, 30 Xue Yuan Road, Beijing , P. R. China low-carbon steel has been reported to date. The reason for this is that the melting point of low-carbon steel is relatively high and its semisolid temperature range is narrow. In this study, the semisolid state behavior and the semisolid extrusion properties of low-carbon steel were investigated focusing on the possibility of clarifying a semisolid forming process. Low-carbon steel, designated as JIS S20C, AISI 1020, BS 070M20 or DIN C22, has applications as a structural material for machine parts, such as bolts, nuts, pins and other structural parts. This material has good forgeability, ease of drawing and is heat treatable; moreover, it is one of the cheapest metal alloys available at present. From the results of this study, it has been observed that there is great potential to obtain various shapes, microstructures and mechanical properties of low-carbon steel directly from molten steel through semisolid-metal forming. 2. Experimental Method and Conditions 2.1 Test material A rolled bar of a commercially available low-carbon steel was used in this study. Table 1 shows its chemical composition. The carbon content is 0.19 wt%; thus, the material is a low-carbon steel. The cooling curve of the low-carbon steel was obtained. The measuring method was as follows. The test piece, which consists of a 32-mm-diameter, 40-mm-length bar, was placed in an alumina crucible in which #1500 alumina powder was filled between the test piece and the crucible. A platinumrhodium thermocouple was attached to the center of the test piece. As can be seen in Fig. 1, the liquidus temperature is 1520 C (1793 K), and the solidus temperature is 1478 C Table 1 Chemical composition of the test material (wt.%). C Si Mn P S Cu Ni Cr

2 808 S. Sugiyama, J. Li and J. Yanagimoto Thermo-couple Billet 1793K(1520 C) Alumina powder 1751K(1478 C) R-type thermocouple Graphite case Graphite block Temperature 1div.: 50K Cooling rate Alumina powder Test material Alumina crucible Pin Time 1div.:50s Induction furnace Fig. 1 Cooling curve of the low-carbon steel. (a) (b) Details of the part indicated in (a) Fig. 3 Schematic illustration of the billet heating. Fig. 2 Microstructure of the low-carbon steel quenched from 1496 C (picric acid etching). (1751 K). The range from the liquidus temperature to the solidus temperature is only 42 K, which is less than half that of cast irons and tool steels. The cooling rates can also be determined from Fig. 1: they are 2.5 K/s in the molten-metal region, 1.8 K/s in the solid region and only 0.3 K/s in the semisolid region. The slow cooling rate observed in the semisolid region is advantageous, because the temperature decrease during the transfer of the billet from the furnace to the forming press is expected to be small. Figure 2 shows the microstructure of the low-carbon steel rapidly quenched from 1496 C (1769 K). The test was carried out using a multistage compression test machine in a nitrogen atmosphere. The specimen is 8 mm in diameter by 12 mm in length. The procedure was as follows. The material was heated using a 100 khz, 20 kw induction furnace, maintained at the test temperature and rapidly quenched to freeze the microstructure. It can be seen that a globular structure with a grain size of mm exists in this semisolid temperature range. It is known that aluminum alloys, magnesium alloys, and cast irons or stainless steels processed in their semisolid ranges have similar structures 15) 2.2 Extrusion method A schematic illustration of the billet heating is shown in Fig. 3. The thermocouple was attached to the central part of the billet. A graphite case and a graphite block surrounded the billet. They were heated in air using a 3 khz, 30 kw induction furnace, which was placed close to a 100 ton press machine equipped with an extrusion unit. During heating, the graphite case, the graphite block and alumina powder were used to prevent a rapid temperature decrease of the billet, to maintain the billet in a semisolid state and to prevent carbon diffusion into the billet. The graphite case was held by a pin in the graphite block. The billet used for the extrusion test was 18 mm in diameter by 35 mm in length and was machined from a hot-rolled bar. Figure 4 shows the temperature decrease from the semisolid temperature of the billet surrounded by the case and block. It takes 47 s for the temperature to decrease from 1505 C (2778 K) to 1475 C (1748 K), which means that the rate of temperature decrease was approximately 0.6 K/s. The carbon block was transferred from the furnace to the forming press in approximately 5 10 s. Thus, the real extrusion

3 Semisolid Extrusion of Low-Carbon Steel 809 Table 2 General conditions of extrusion test. Fig. 4 Temperature decrease of billet surrounded by case and block. Billet Material JIS S20C (=AISI 1020) Size (Diameter-Height) /mm Semisolid Temperature Range / C (K) ( ) Extrusion Container size (Diameter-Length) /mm Die Size (Diameter-Land Length) /mm 7 8 Shape Flat Material JIS SKD61 (AISI H13) Extrusion Temperature / C (K) ( ) Extrusion Ratio 10 Ram speed / mms Lubricant Dry Forced Cooling Compressed air, Mist water spray Punch Container Billet Graphite case Hot extrusion Die Gus, mist, water inlet Die holder Fig. 6 Extrusion force (P) vs stroke (S) curves tested at C. Fig. 5 Schematic illustration of the extrusion test. temperature of the billet was assumed to be 3 6 K less than the measured temperature. The extrusion test was carried out by the following steps. (a) Once the target temperature is reached, the graphite block is manually taken out of the induction furnace. (b) The graphite block is placed on the top of a container. (c) The pin is pulled out and the graphite case containing the billet is dropped into the container. (d) A punch is set on the graphite case. (e) The extrusion process is initiated. Depending on the experimental conditions, compressed air or water mist is blown onto the extruded product at the die exit. The punch and the container are kept at room temperature before extrusion. Figure 5 shows a schematic illustration of the extrusion test. A summary of the general conditions for extrusion are shown in Table Experimental Results and Discussion 3.1 Extrusion force vs stroke curve Figure 6 shows the relationship between extrusion force and ram stroke. The extrusion temperatures of 1304 C (1577 K) and 1403 C (1676 K) are in the range of conventional hot working. The workpiece was in the semisolid extrusion range at 1480 C (1753 K) and 1490 C (1763 K). Each curve shows three stages; i.e., the load increases at a high rate in the early stage of extrusion (stage 1), at moderate rate (stage 2) and again at a high rate (stage 3). This behavior can easily be explained considering that the billet fills the container during stage 1, is extruded in stage 2, and is still being extruded in stage 3, but the load increase accounts for the billet temperature decrease. By comparison with the extrusion force at stage 2, note that the hot extrusion process requires approximately twice the force of the semisolid extrusion force. In stage 1 of the semisolid extrusion, the surplus liquid component in the billet moves much more rapidly than the solid component to the die exit under a relatively small pressure. The same results have been observed in the semisolid extrusion of aluminum alloys. 1) 3.2 Surface quality Figure 7 shows the appearance of the extruded samples and close-ups of their top part. It is observed that hotextruded samples have a better-quality surface along the whole length of the bar. In contrast, the semisolid-extruded samples have a poor quality near the top. At the beginning of the process in the semisolid state, the liquid component is

4 810 S. Sugiyama, J. Li and J. Yanagimoto (a) Hot extrusion product is more pronounced in comparison with the nonforced cooling product. The multilayered structure continues from the top to the bottom of the product. (b) Hot extrusion Fig. 7 Appearance of the samples and close-up of their top part extruded in solid and semisolid states. expelled, then a high-quality surface is obtained as the extrusion proceeds. 3.3 Cross-sectional observation Figure 8 shows the results of macroscopic observation of the cross-section of the extruded products. The sampling position is about 1/3 of the way from the top of the extruded product. The cross section of the product of hot extrusion is shown in Fig. 8(a), and the semisolid-extruded products are shown in Figs. 8(b) and (c). The extruded product undergoing forced cooling (mist cooling) at the die exit is shown in Fig. 8(c), and the products undergoing nonforced cooling (natural cooling) are shown in Figs. 8(a) and (b). A multilayered structure in the semisolid-extruded products can be seen, and the multilayered structure of the forced-cooling 3.4 Microstructures Figure 9 shows the microstructures of the as-extruded bars. The upper pictures show the periphery of the crosssection and the lower ones show the center. In the case of hot extrusion, even though the center has a much larger grain size than the periphery, the grain shapes are similar throughout the cross section. The grain size is approximately mm. In the hot-extruded product, the white parts in the picture are ferrite, and the gray parts are perlite. On the other hand, the semisolid-extruded products have a completely different structure. They have a trilaminar structure. The outer layer is an aggregate of mm grains, the middle layer is an aggregate of mm grains, and the inner layer (core part) is an aggregate of mm grains. It was verified that the longitudinal and cross-sectional structures of the extruded bars are almost the same. The phenomena observed in the semisolid-state extrusion are very complicated; they are the result of the interaction of variables such as extrusion temperature, cooling rate, strain and strain rate, which may result in oxidization or decarburization, recrystallization and phase transformation during cooling. To understand these complex phenomena, much more research and consideration is required. 3.5 Componential analysis Figure 10 shows the energy-dispersive X-ray spectroscopy (EDS) results of the extruded samples. An X-ray line analysis of C, Si, Mn and S elements in the radial direction of the cross section was carried out for the extruded samples. It can be seen that the carbon content of the periphery of the semisolidextruded sample is clearly lower than that of the centre. This phenomenon occurs for two reasons. First a material flow difference occurs between the liquid component and the solid component under semisolid extrusion. The liquid component collects at the center, since it has a higher fluidity than the solid component. Second, the carbon content is mainly (a) (b) (c) Fig. 8 Appearance of the cross sections of the extruded samples (picric acid etching). (a) Hot extrusion with natural cooling, (b) Semisolid extrusion with natural cooling and (c) with mist cooling.

5 Semisolid Extrusion of Low-Carbon Steel mm 0.1mm 0.1mm 0.1mm 0.1mm 0.1mm (a) (b) (c) Fig. 9 Microstructures of the extruded samples of low-carbon steel (picric acid etching). Upper pictures: cross-section at the periphery. Lower pictures: cross-section at the center. (a) Hot extrusion with natural cooling, (b) with natural cooling and (c) with mist cooling. (a) (b) C Si Mn S Fig. 10 Distributions of elements in cross-section of the extruded products. (a) Semisolid-extruded sample. (b) Hot-extruded sample. contained in the liquid component. Other elements, such as Si and Mn, have slightly higher concentrations at the periphery than at the central portion. In the case of the hot-extruded samples, C, Si, Mn and S have an almost constant concentration from the centre to the periphery. 3.6 Hardness of extruded products Figure 11 shows the distributions of Vickers hardness along the radial direction along the cross section of the products extruded under both hot and semisolid states under various cooling conditions. As shown in the figure, the hardness of the hot-extruded bars is almost fixed across the cross section, regardless of the cooling conditions. On the other hand, the hardness of the semisolid-extruded bars differs markedly from the hot-extruded bar. The periphery of the semisolid-extruded products is soft, the hardness is nearly

6 812 S. Sugiyama, J. Li and J. Yanagimoto Vickers Hardness HV (1) Cross Section Fig. 11 equal to that of the hot-extruded product, but the centre is about two times harder. This is because the microstructure in the center is different from the microstructure of the periphery when the carbon content concentrates in the centre. 4. Conclusions 1577K Natural Cooling (N.C.) 1575K Mist Cooling (M.C.) 1771K N.C. 1763K M.C. Hot Extrusion Location of Measurement, L/mm Hardness test results of the extruded products. Semisolid processing of aluminum alloys or magnesium alloys has recently become common and many investigations on this process have been reported, whereas there have been few applications of semisolid processing to higher-meltingpoint alloys, such as low-carbon steel. It is clear from our study that (1) low-carbon steel can be processed by semisolid processing, (2) the extrusion force is approximately half that of hot forming, (3) the chemical composition of carbon in semisolid-extruded samples is higher at the centre, whereas silicon and manganese have slightly higher concentrations at the periphery, and (4) the microstructure and mechanical properties such as hardness exhibit marked changes in the radial direction. The semisolid extrusion of low-carbon steel is a possible first step towards achieving a high rate of productivity for a small-forming unit of low-carbon steel with an extremely fine and gradient-function microstructure. REFERENCES 1) M. Kiuchi, S. Sugiyama and K. Arai: The 10th M.T.D.R. (Machine Tool Design and Research), 8 (1979) ) P. Kapranos, D.H. Kirkwood and C. M. Sellars: The 2nd Int l Conf. on Semi-Solid Processing of Alloys and Composites, 6 (1992) ) S. P. Midson, N. H. Nicholas, R. A. Nichting and K. P. Yong: The 2nd Int l Conf. on Semi-Solid Processing of Alloys and Composites, 6 (1992) ) M. Kiuchi, S. Sugiyama and M. Arai: Journal of JSTP, (1996) ) B. Nohn, U. Morjan and D. Hartmann: The 6th Int l Conf. on Semi- Solid Processing of Alloys and Composites, 9 (2000) ) R. Kopp, J. Kallweit and Th. Moller: The 6th Int l Conf. on Semi-Solid Processing of Alloys and Composites, 9 (2000) ) M. Kiuchi and S. Sugiyama: The 1st Int l Conf. on Semi-Solid Processing of Alloys and Composites, 4 (1990). 8) N. Yoshida, Y. Murata, T. Moriya and C. Yoshida: The 3rd Int l Conf. on Semi-Solid Processing of Alloys and Composites, 6 (1994) ) Y. Kang, R. Song, J. Sun, A. Zhao, J. Wang and W. Mao: The 7th Int l Conf. on Semi-Solid Processing of Alloys and Composites, 9 (2002) ) P. Secordel, E. Valette and F. Leroy: The 2nd Int l Conf. on Semi-Solid Processing of Alloys and Composites, 6 (1992) ) K. Miwa and S. Kawamura: The 6th Int l Conf. on Semi-Solid Processing of Alloys and Composites, 9 (2000) ) S. Abdelfattah, M. Robelet, A. Rassili and M. Bobadilla: The 6th Int l Conf. on Semi-Solid Processing of Alloys and Composites, 9 (2000) ) C. Rouff, R. Bigot, V. Favier and M. Robelet: The 7th Int l Conf. on Semi-Solid Processing of Alloys and Composites, 9 (2002) ) M. Robelet, J. Demurger, A. Rassili, D. Fischer, H. Klemm, B. Walkin, M. Kalsson and A. Cucatto: The 8th Int l Conf. on Semi-Solid Processing of Alloys and Composites, 9 (2004). 15) S. Okano and M. Kiuchi: The 7th Int l Conf. on Semi-Solid Processing of Alloys and Composites, 9 (2002)