- Email: [email protected]
Solid-state recycling of aluminium alloy swarf into c-channel by hot extrusion
Journal of Manufacturing Processes 17 (2015) 1–8
Contents lists available at ScienceDirect
Journal of Manufacturing Processes journal homepage: www.elsevier.com/locate/manpro
Technical Paper
Solid-state recycling of aluminium alloy swarf into c-channel by hot extrusion Ryoichi Chiba ∗ , Morihiro Yoshimura 1 Department of Mechanical Systems Engineering, Asahikawa National College of Technology, 2-2-1-6 Shunkodai, Asahikawa, Hokkaido 071-8142, Japan
a r t i c l e
i n f o
Article history: Received 3 October 2013 Received in revised form 1 October 2014 Accepted 15 October 2014 Keywords: Aluminium alloy C-channel Extrusion Machining swarf Mechanical property Solid-state recycling
a b s t r a c t In this study, we have investigated the possibility of the solid-state recycling of aluminium alloy machining swarf into c-channels using hot extrusion. Side milling swarf and lathe turning swarf generated from a cast Al–Si alloy ingot were cold compacted into columnar billets and successfully profile-extruded into equilateral c-channels at 600 K, under extrusion ratios of 10 and 18. The c-channels obtained at an extrusion ratio of 18 showed straight extrusion without warping, except in the front-end region. In case of the material recycled from the milling swarf at an extrusion ratio of 10, the optical microscopic study indicates the presence of coarse residual voids and cracks existing in regions where sufficient plastic strain was not introduced. In contrast, the material recycled from the same swarf at an extrusion ratio of 18 did not contain any coarse voids, rather had a density comparable to that of the original ingot. This could be attributed to the large strain of over 4.3 that was introduced in the sample recycled at the extrusion ratio of 18, as predicted by finite element analysis. Uniaxial tensile tests showed that the dense recycled material had a higher ductility than the original ingot, with a reduction of around 30% in the ultimate tensile strength. The material recycled from the turning swarf exhibited insufficient bonding among the individual pieces of the swarf, as compared to that recycled from the milling swarf under the same conditions, thereby resulting in inferior mechanical properties. © 2014 The Society of Manufacturing Engineers. Published by Elsevier Ltd. All rights reserved.
1. Introduction Currently, most aluminium scrap, including machining swarf, i.e., small chips, is remelted in furnaces and recycled into ingots or die-casting products. However, the conventional recycling method is associated with the following disadvantages: (1) the metal yield rate is very low, i.e., approximately 55% [1]. (2) Unavoidable reduction in the purity of recycled ingots at the remelting stage results in the degradation of their mechanical properties. (3) The remelting process requires significantly large amount of energy, which is unfavourable in terms of energy conservation. These limitations drive the development of alternative technologies for the recycling of aluminium scrap. Recently, solid-state recycling, which involves direct recycling of metal scrap into bulk material by using severe plastic deformation (SPD), has emerged as a potential alternative to the conventional remelting and recycling techniques [2]. The
∗ Corresponding author. Tel.: +81 166 55 8003; fax: +81 166 55 8003. E-mail address: [email protected] (R. Chiba). 1 Present address: Engineering Division, Oji Materia Co., Ltd., 20-6 Tokuda, Nayoro, Hokkaido 096-8555, Japan.
solid-state recycling method not only reduces the energy consumption but also improves the recycling efficiency dramatically (up to more than 95% for aluminium [1]). In addition, the solid-state recycling method offers the advantage of microstructural control during the recycling process, which allows the metal scrap to be recycled into materials with excellent mechanical properties. Thus far, several solid-state recycling processes have been intensively studied since 1980s, especially to recycle the machining swarf discharged from factory machine tools. A great majority of studies on this topic have focussed on recycling machining swarf into cylindrical bars [3–12] or rectangular bars [13–15], using hot/warm forward extrusion. In addition, recycling processes based on other SPD methods such as cyclic extrusion compression (CEC) [16], equal channel angular pressing (ECAP) [17–19], high-pressure torsion (HPT) [20] or compressive torsion [21], and combined use of forward extrusion and ECAP [22,23] have also been reported. Recently, another study [24] aimed at producing a composite material including nano-particles from machining chips via hot extrusion, and a new method was proposed to produce clad plates from iron chips by using hot rolling [25]. Almost all of the abovementioned methods require precompaction of the machining swarf, the technical details of which have been reported elsewhere [26]. A brief review of the relevant
http://dx.doi.org/10.1016/j.jmapro.2014.10.002 1526-6125/© 2014 The Society of Manufacturing Engineers. Published by Elsevier Ltd. All rights reserved.
2
R. Chiba, M. Yoshimura / Journal of Manufacturing Processes 17 (2015) 1–8
Table 1 Chemical composition of the AC4CH aluminium alloy used in this study [27]. Element
Cu
Si
Mg
Zn
Fe
Mn
Ti
Sb
Al
Mass%
0.01
6.9
0.37
0.02
0.13
0.01
0.13
0.001
Bal.
literature published before 2010 can be found in our previous paper [27]. Extruded sections are elongated metal products that have a complex configuration of the cross section, represented by angles and channels. Their cross-sectional shapes are non-axisymmetric and more complex than those of simple-shaped bars, e.g., cylindrical bars and rectangular bars. Extruded sections are in high demand in the industry, similar to the simple-shaped bars and metal plates. In recent years, the demand for extruded sections of aluminium alloy has been significantly increasing with the growing needs of applications demanding reduction in weight. Therefore, it is highly necessary to develop methodologies for recycling metal scrap into extruded sections. This is expected to increase the opportunities of using recycled materials, thereby leading to the promotion of use of recycled resources. However, to the best of the authors’ knowledge, there are no attempts reported so far on the direct solid-state recycling of machining swarf into extruded sections. Extrusion to form complex configuration of the cross section is different in several ways from extrusion into simple cylindrical/rectangular bars. First, if the complex configuration is asymmetric, the extrudate may exhibit warping. Second, the streamlines of the plastic flow are likely to change steeply and may bifurcate. Third, because the slit width of dies can be location dependent, there exist portions that undergo deformation for a local extrusion ratio considerably larger than the global extrusion ratio. Fourth, complex configuration generally has many corners and the contact area (friction area) of die land is very large compared to the case of simple configuration for the same extrusion ratio. In other words, a wide area undergoes significant shear deformation. Of course, the performance of the solid-state recycling is affected by the above factors. The present study investigates the possibility of recycling aluminium alloy machining swarf into c-channels using hot extrusion. Two types of machining swarfs of different shapes were coldcompacted into the shape of a billet, and subsequently respective billets were made into equilateral c-channels through hot profile extrusion process at different extrusion ratios. The surface appearance, density, microstructure, and mechanical properties of different specimens obtained under different recycling conditions were compared. Finally, we have discussed the feasibility of adopting cold-compaction followed by hot extrusion, for the solid-state recycling of aluminium alloy machining swarf into c-channels. 2. Experimental procedure
Fig. 1. Appearance of the machining swarf obtained by (a) side milling (reproduced from Fig. 1a in [27]) and (b) lathe turning.
obtained by lathe turning is found to be curled, short swarf of length 5–30 mm, width 1.5 mm, and thickness 0.5 mm. The machining swarf was ultrasonically degreased with acetone for 10 min. The cleaned machining swarf was then placed in a cylindrical container and compacted at a pressure of 303 MPa [27,28] under room temperature to form billets of diameter 20.5 mm and height approximately 25 mm. The compacts were unloaded immediately after the loading pressure reached 303 MPa. A universal testing machine (Autograph AG-X 250 kN, Shimadzu) was used for the compaction as well as for the subsequent extrusion process, described in the following section. As a pre-processing step, the compacts (billets) were annealed at 300 ◦ C for 1 h, before the extrusion process. This was performed to prevent cracks in the extrudates, which were observed in the hot extrusion of the ascompacted billets.
2.1. Machining swarf production and cold-compaction 2.2. Hot extrusion Two types of machining swarfs with different shapes were produced by side milling and turning operations of a commercial AC4CH aluminium alloy ingot with cutting oil. The chemical composition of the aluminium alloy used in this study is summarized in Table 1. In case of side milling, the cutting conditions, namely, cutting rate, feed rate, and cut depth, adopted during the machining operations were 46 m/min, 0.1 m/min, and 1 mm, respectively, while those for turning were 67 m/min, 0.1 mm/rev, and 0.5 mm, respectively. Fig. 1 shows the appearance of the machining swarf. As can be seen from the figure, the machining swarf obtained by side milling is needle-shaped and spirally twisted of length 25–30 mm and width 1 mm [27]. On the other hand, the machining swarf
In the present study, the extrusion was performed by the forward extrusion method. The dies used in this study have a C-shaped orifice at the centre (Fig. 2), so that a symmetry axis passes centrally in the dies. The dimensions x and y were kept as the same, in order to maintain the ratio between the height and width of the c-channel products as 1:1. In addition, the corner radius of 0.5 mm was set at all the eight corners. An approach region with a die angle of 45◦ was also set to get a smooth transition from the end of the container to the inlet of the dies. Two types of dies with different extrusion ratios R = 10 and 18 were prepared, which have an identical die land length of 3 mm.
R. Chiba, M. Yoshimura / Journal of Manufacturing Processes 17 (2015) 1–8
> 40 mm
3
30 mm Tab 1 mm
t
Grip section
A
A
Fig. 4. Edge-view sketch of tabbed tensile specimens.
O
y
t
E
x
x = y = 9.1, t = 1.3 and E = 4.55 for R = 10
B
x = y = 7.0, t = 1.0 and E = 3.5 for R = 18 (a) A—O—A A— O — B
Gauge section
R 0.5
Die angle 45°
t
1 (Approach) 3 (Die land)
30° Back relief angle (b) Fig. 2. Dimensions (in mm) of orifice on (a) die face and (b) cross sections along the thickness direction.
The annealed compact was placed in a cylindrical container, and then extruded into c-channels at 600 K using the universal testing machine. The temperature during the extrusion process was controlled by using a mantle heater (Tokyo Garasu Kikai Co. Ltd.) installed around the container and a desktop temperature adjuster (TC-1N, AS-ONE). Also, a MoS2 -based lubricant was applied over the inner surface of the container, and the taper and land surfaces of the dies at every extrusion, in order to prevent the increase in the extrusion load as a result of friction. The ram speed was maintained at 5 mm/min, resulting in an extrusion velocity of ∼1.5 mm/s. The compacts made of the milling swarf were extruded at R = 10 and 18, while those made of the turning swarf were extruded only at R = 18. 2.3. Testing of mechanical properties and microscopic observation Tensile specimens were cut out from the steady deformation region of the extrudates in the form of c-channels. Specifically, the extrudates were cut into side parts and a back part (Fig. 3), and each part was subjected to uniaxial tension parallel to the extrusion direction (ED), in order to evaluate the 0.2% proof stress, ultimate
tensile strength (UTS), and uniform elongation. Because the small width of the cut parts made it difficult to machine a reduced gauge section, the cut parts were tabbed prior to the testing to avoid unwanted failure in the grip sections, as shown in Fig. 4. That is, 1mm-thick AA1100 aluminium alloy tabs were bonded onto both the sides with Araldite. The longitudinal strain was measured using a strain gauge adhered to the centre of the gauge section surface. The tensile tests were conducted at room temperature at a crosshead speed of 5 mm/min on the universal testing machine. All the results reported in this study are an average of at least three tensile tests. Additionally, Vickers hardness test was performed at an applied load of 9.8 N and a holding time of 15 s using a micro-hardness tester (model-MVK, Akashi). The microstructure of the recycled materials was observed by using optical microscopy. Prior to the observation, the surfaces of the specimens were polished with emery paper and buffed with an alumina suspension. Subsequently, the specimens were etched using 0.5 ml of 40% hydrofluoric acid and 100 ml of distilled water. 2.4. Finite element analysis of extrusion process Numerical analyses of the extrusion processes were performed using a commercial finite element code DEFORM-3DTM , in order to investigate the extrusion-induced strain distribution in the transverse section (i.e., the plane perpendicular to the ED) of the extruded c-channels. Considering the symmetry of the deformation, a half of the billet was meshed with 32,000 linear tetrahedral elements. In particular, the mesh in the die inlet was maintained fine in order to resolve the large shearing of the material that takes place in this region. An isotropic rigid perfectly plastic model obeying von Mises yield criterion was used for the billet (i.e., the billet was assumed to be void-free and the density of billet was taken as the theoretical density of aluminium [15]), whereas the tools were defined to be rigid bodies. The experimental flow stress data obtained from the uniaxial compression tests of the annealed AC4CH ingot at an initial strain rate of 0.014 s−1 at 600 K were used as the deformation resistance properties in the constitutive model. The experimental stress–strain curve is presented in Fig. 5, which shows the same level of flow stress as Al–12Si eutectic alloy [29]. The friction between the billet and tools was considered to obey the Coulomb rule, with an assumed constant friction coefficient of 0.2. 3. Results and discussion 3.1. Microstructure of machining swarf and billet density
Fig. 3. Cutting of c-channel material for tensile specimens.
The microstructures of the original AC4CH ingot and milling swarf are shown in Fig. 6. The microstructure of the original ingot reveals needle-like eutectic Si phases in coarse primary Al phases. Moreover, it could be observed that the Si phases were broken into short needle-like particles of average length ∼15 m as a result of the cutting work. As described above, two types of machining swarfs were pressed to form billets of diameter 20.5 mm. The density of the billets made of milling swarf and turning swarf was found to be 2.397 g/cm3 and 2.386 g/cm3 , respectively, which correspond to relative densities of
4
R. Chiba, M. Yoshimura / Journal of Manufacturing Processes 17 (2015) 1–8
Fig. 5. Experimental and approximated stress–strain curves of annealed AC4CH ingot.
0.891 and 0.887, respectively. This indicates that the billets contain many voids. Furthermore, the macroscopic inspection of the surfaces of both the billets revealed that the size of an individual void was larger in case of billet derived from the turning swarf.
Fig. 7. Appearances of the extrudates, as seen from (a) lateral view and (b), (c) bird view.
3.2. External appearance of the recycled materials
Fig. 6. Optical micrographs of (a) original AC4CH ingot (reproduced from Fig. 9a in [27]) and (b) machining swarf obtained by side milling.
Fig. 7 shows the external appearance of the specimens obtained through the hot extrusion processes of the billets made of the milling swarf (Fig. 1a). For comparison, specimens extruded without lubrication are also shown in the figure. The surfaces of the specimens extruded with lubricant were found to be shiny and smooth for both R = 10 and 18, while those extruded without lubricant were not shiny and smooth because of the considerable friction against the dies. In some specimens extruded at R = 10, saw bladelike tears were observed at the tip of the side parts, as shown in Fig. 7a. The specimens extruded at R = 18 exhibited warping in the region less than about 30 mm away from the front end due to the end effect, but showed straight extrusion in the other region. In
R. Chiba, M. Yoshimura / Journal of Manufacturing Processes 17 (2015) 1–8
5
Fig. 8. Optical micrographs of specimens extruded from milling-swarf billets at (a), (b) R = 10; (c) R = 18; and (d), (e) extruded from a turning-swarf billet at R = 18.
contrast, the specimen extruded at R = 10 with lubrication exhibited warping throughout the length of the specimen. The surfaces of the specimens recycled from the turning swarf (Fig. 1b) had the same characteristics as those of the recycled specimens obtained from the milling swarf. Note that the authors considered only the specimens that were extruded with lubrication as a target for the following discussion. 3.3. Macro- and micro-structures of the recycled materials Fig. 8 shows the optical micrographs of the transverse section of the recycled specimens (extrudates). As can be seen from the optical images shown in Fig. 8a and b, the material recycled from the milling swarf by hot extrusion at R = 10 exhibited relatively large voids and cracks. This indicates that the bonding among the individual pieces of swarf is insufficient. On the other hand, the material recycled from the same swarf by extrusion at R = 18 had no observable cracks located at the bonded interfaces of the swarf. Therefore, the corresponding optical image revealed a highly dense surface, although minute (<5 m) voids were slightly observed. Furthermore, the optical image of the material recycled from the turning swarf by extrusion at R = 18 (Fig. 8d and e) revealed a number of voids and long cracks along the swarf interfaces in the corner parts and adjacent to the surfaces. The cracks may be attributed to the inability of the turning swarf billets to accommodate large shear
strain induced by extrusion. That is, the milling swarf, which is thinner than the turning swarf, has a higher deformability and can gain entry into tiny spaces to readily form a continuum body without voids. On the other hand, the individual voids originally included in the turning swarf billets are larger than those in the milling swarf billets, despite their comparable densities. Therefore, even an extrusion at R = 18 is not sufficient to collapse the large voids. Hence, the severe shear deformation of the material, including the residual voids near the surfaces, causes obvious cracks along the swarf interfaces (Fig. 8d and e). Fig. 9 shows the microstructure of the extruded specimens at the centre of the transverse section of the back part (see Fig. 3). As evidenced from the figure, no boundaries between the chips were found in the materials recycled from the milling swarf, whereas the boundaries were clearly observed in the material recycled from the turning swarf. It can also be seen that the eutectic Si phases were refined, when compared with those in the original ingot and machining swarf. There was no large difference in the size of Si phases between the materials recycled at different extrusion ratios (R = 10 and 18) or between the materials recycled from the differently shaped swarfs; the Si phases/particles of every recycled material ranged from 1 to 10 m in size. However, comparing Fig. 9a and b, it appears that the Si particles are more isolated in the materials recycled at R = 18, when compared with the material recycled at R = 10.
6
R. Chiba, M. Yoshimura / Journal of Manufacturing Processes 17 (2015) 1–8 Table 2 Relative density of the materials recycled from milling swarf. Billet
Relative density
0.891
Extrudates R = 10
R = 18
0.970
0.998
Fig. 10. Mechanical properties of each separated part of the recycled material in the form of c-channel.
3.4. Relative density With reference to the density of the AC4CH ingot, the relative densities of the materials recycled from the milling swarf were measured using the Archimedes method. The corresponding results shown in Table 2 imply that the specimen extruded at R = 10 still has considerable voids. This corresponds to the fact that the residual voids and cracks were observed in the specimen, as shown in Fig. 8a and b. The density of the material recycled at R = 18 was nearly identical to that of the original ingot. 3.5. Mechanical properties at room temperature
Fig. 9. Microstructure of the specimens extruded from milling-swarf billets at (a) R = 10 and (b) R = 18 (c) extruded from a turning-swarf billet at R = 18.
Fig. 10 shows the tensile test results (0.2% proof stress, UTS, and uniform elongation) of the separated parts of the respective extruded specimens. The tensile test results indicate greater variability in the uniform elongation than in the proof stress and UTS, similar to the case of solid-state recycled magnesium alloy [6]. Furthermore, major differences in the strength level and ductility between the side part and back part could be observed in all the obtained c-channels. Both the strength level and ductility were found to be higher in the side part. This could be attributed to the difference in the strain introduced during extrusion processes, as described in the forthcoming section. The 0.2% proof stress, UTS, and fracture strain of the AC4CH ingot were 133.5 MPa, 191.1 MPa, and 0.030, respectively (details not shown in Fig. 10) [27]. Thus, all the recycled materials exhibited a decrease in the mechanical strengths. However, the material recycled from the milling swarf by extrusion at R = 18 showed an increase in the ductility. The observed improvement in ductility is due to the Si-phase refinement, sufficient void reduction, and probably the partial dynamic recrystallization. Since the aluminium alloy extruded sections generally undergo secondary processing, such as bending, to form actual structural parts, their ductility is a very important factor from the application perspective [30]. The material recycled by extrusion at R = 10 had lower ductility. Some of the side parts cut out from
R. Chiba, M. Yoshimura / Journal of Manufacturing Processes 17 (2015) 1–8
7
Table 3 Vickers hardness of the materials recycled from milling swarf. Extrusion ratio
Side part
Back part
R = 10 R = 18
44.7 HV(1.0) 49.4 HV(1.0)
43.9 HV(1.0) 49.6 HV(1.0)
this recycled material fractured with little plastic deformation. This indicates that the bonding among the individual pieces of swarf is imperfect (Fig. 8a and b). However, decrease in the mechanical strength of recycled materials is not a critical disadvantage, as several applications (such as window frames) do not demand the full strength and ductility of as-cast aluminium, allowing solid-state recycled materials to be used instead [2]. Next, the effect of the shape of machining swarf on the mechanical properties of the recycled materials is discussed. As can be seen from Fig. 10, both the mechanical strength and uniform elongation of the specimen recycled from the turning swarf are inferior to that from the milling swarf. This is the synergetic effect of residual chip boundaries (shown in Fig. 9c) and a number of voids and cracks (Fig. 8d and e). The milling swarf and turning swarf differ markedly in the ratio of surface area to volume, i.e., contamination level of oxides. However, their effects on the mechanical properties of the recycled materials are minor, as the differences in the contamination of oxides have negligible effects on the mechanical strength of recycled aluminium materials at a wide range of temperature and on the ductility at room temperature [31]. As stated earlier, the main reason underlying the observed poor mechanical properties is the presence of large voids originally included in the turning swarf billets, which were not sufficiently collapsed during the extrusion process. One way to improve the bonding between the chips and reduce the residual voids and cracks in the specimen recycled from the turning swarf is to increase the extrusion temperature and decrease the extrusion rate. Gronostajski et al. [32] states that a high extrusion temperature makes the plastic flow of the material possible into pores and voids, and that a relatively low extrusion rate gives the time necessary for the diffusional transport of matter. Vickers hardness was measured at two different points on the transverse section of the specimens recycled from the milling swarf. The measurement locations are the centre of the transverse section of each separated part. The measured results are summarized in Table 3. The effect of measurement location on the hardness was very small, irrespective of the extrusion ratio. On the other hand, considering the effect of extrusion ratio, the material recycled by extrusion at R = 18 is found to be harder in both the parts. This result corresponds to the trend observed in the mechanical strength of the specimens (Fig. 10). 3.6. Numerical prediction of introduced strain Fig. 11 shows the distribution of equivalent strain in the cchannels extruded at R = 10 and 18, as predicted through the finite element analyses. For each extrudate, the contour plot is displayed in the transverse section of the steady deformation region. It can be observed that both the c-channels have strain non-uniformly distributed in the cross section and a large strain is introduced at corners. Moreover, in each extrusion case, the overall strain level is higher in the side parts than in the back part. This would lead to higher strength and ductility of the side parts in the tensile tests. A comparison of the extrudates obtained by extrusion at R = 10 and 18 presents a difference of 0.5 in the minimum value of equivalent strain. This difference improved the bonding state among individual pieces of machining swarf in the material recycled at R = 18, resulting in higher mechanical strength and elongation. Note that because the minimum equivalent (plastic) strain is given by ln R for
Fig. 11. Spatial distribution of equivalent strain in specimens extruded at R = 10 and 18. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
extrusion into simple cylindrical bars, those strain values are calculated to be 2.3 for R = 10 and 2.9 for R = 18, which are lower than the minimum values in Fig. 11 by about 1.5. As can be seen from Figs. 10 and 11, the mechanical properties of the recycled materials are positively correlated with the introduced strain. 4. Conclusions Recycled materials in the form of equilateral c-channel were obtained from the milling and turning swarf of Al–Si alloy (AC4CH) by cold-compaction followed by hot extrusion at 600 K at two different extrusion ratios of R = 10 and 18. The following conclusions can be drawn on the basis of the systematic investigation of their microstructure and mechanical properties at room temperature: (1) The materials recycled from the milling swarf via hot extrusion process at two different extrusion ratios had lower proof stress and ultimate tensile strength than those of the original ingot. However, the material recycled at an extrusion ratio of 18 exhibited an increase in the ductility because of the fine dispersion of Si particles with sufficient void reduction. (2) The mechanical strength and uniform elongation of the recycled materials were both positively correlated with the plastic strain introduced into the c-channels during the extrusion. Thus, in terms of the mechanical properties, the side parts of the channels tend to be superior to the back parts. (3) The material recycled from the turning swarf had many voids and cracks, and also chip boundaries, which resulted in inferior mechanical strength and ductility. In contrast, these defects were not included in the material recycled from the milling swarf under the same condition. This suggests that the type of machining swarf significantly affects the mechanical properties of the solid-state recycled materials. (4) The hot profile extrusion of the aluminium alloy machining swarf for an extrusion ratio of 18 or more, corresponding to an introduced equivalent strain of at least 4.3, have good potential
8
R. Chiba, M. Yoshimura / Journal of Manufacturing Processes 17 (2015) 1–8
to produce c-channels of recycled materials with comparable density and superior ductility to the original ingot. They could be potentially used in certain branches of industry, for instance, in applications such as the non-load-bearing components that do not demand the full strength of as-cast aluminium. Acknowledgement The authors would like to thank Mr. Toshiyuki Suzuki and Mr. Toshiyuki Akimoto for preparing the extrusion dies. References [1] Lazzaro G, Atzori C. Recycling of aluminium trimmings by conform process. Light Metals 1992;3:1379–84. [2] Allwood JM, Cullen JM, Cooper DR, Milford RL, Patel ACH, Carruth MA, et al. Conserving our metal energy. WellMet2050. Cambridge: University of Cambridge; 2010. [3] Wu SY, Ji ZS, Rong SF, Hu ML. Microstructure and mechanical properties of AZ31B magnesium alloy prepared by solid-state recycling process from chips. Trans Nonferrous Met Soc China 2010;20:783–8. [4] Anilchandra AR, Surappa MK. Influence of tool rake angle on the quality of pure magnesium chip-consolidated product. J Mater Process Technol 2010;210:423–8. [5] Peng T, Wang Q, Han Y, Zheng J, Guo W. Microstructure and high tensile strength of Mg–10Gd–2Y–0.5Zr alloy by solid-state recycling. Mater Sci Eng A 2010;528:715–20. [6] Zhao ZD, Chen Q, Yang L, Shu DY, Zhao ZX. Microstructure and mechanical properties of Mg–Zn–Y–Zr alloy prepared by solid state recycling. Trans Nonferrous Met Soc China 2011;21:265–71. [7] Zhang T, Ji Z, Wu S. Effect of extrusion ratio on mechanical and corrosion properties of AZ31B alloys prepared by a solid recycling process. Mater Des 2011;32:2742–8. [8] Hu ML, Ji ZS, Chen XY, Wang QD, Ding WJ. Solid-state recycling of AZ91D magnesium alloy chips. Trans Nonferrous Met Soc China 2012;22:s68–73. [9] Anilchandra AR, Surappa MK. Microstructure and damping behaviour of consolidated magnesium chips. Mater Sci Eng A 2012;542:94–103. [10] Miao J, Ye B, Wang Q, Peng T. Mechanical properties and corrosion resistance of Mg–10Gd–2Y–0.5Zr alloy by hot extrusion solid-state recycling. J Alloys Compd 2013;561:184–92. [11] Anilchandra AR, Surappa MK. Microstructure and tensile properties of consolidated magnesium chips. Mater Sci Eng A 2013;560:759–66. [12] Mindivan H, Taskin N, Kayali ES. Recycling of pure magnesium chips by cold press and hot extrusion processes. Acta Phys Pol A 2014;125:429–31. [13] Guley V, Ben Khalifa N, Tekkaya AE. Direct recycling of 1050 aluminum alloy scrap material mixed with 6060 aluminum alloy chips by hot extrusion. Int J Mater Form 2010;3:853–6. [14] Li DH, Hu ML, Wang HB, Zhao WA. Low temperature mechanical property of AZ91D magnesium alloy fabricated by solid recycling process from recycled scraps. Trans Nonferrous Met Soc China 2011;21:1234–40.
[15] Guley V, Guzel A, Jager A, Ben Khalifa N, Tekkaya AE, Misiolek WZ. Effect of die design on the welding quality during solid state recycling of AA6060 chips by hot extrusion. Mater Sci Eng A 2013;574:163–75. [16] Peng T, Wang QD, Han YK, Zheng J, Guo W. Consolidation behavior of Mg–10Gd–2Y–0.5Zr chips during solid-state recycling. J Alloys Compd 2010;503:253–9. [17] Misiolek WZ, Haase M, Ben Khalifa N, Tekkaya AE, Kleiner M. High quality extrudates from aluminum chips by new billet compaction and deformation routes. CIRP Ann – Manuf Technol 2012;61:239–42. [18] Luo P, McDonald DT, Zhu SM, Palanisamy S, Dargusch MS, Xia K. Analysis of microstructure and strengthening in pure titanium recycled from machining chips by equal channel angular pressing using electron backscatter diffraction. Mater Sci Eng A 2012;538:252–8. [19] Luo P, McDonald DT, Palanisamy S, Dargusch MS, Xia K. Ultrafine-grained pure Ti recycled by equal channel angular pressing with high strength and good ductility. J Mater Process Technol 2013;213:469–76. [20] Abd El Aal MI, Yoo Yoon E, Seop Kim H. Recycling of AlSi8Cu3 alloy chips via high pressure torsion. Mater Sci Eng A 2013;560:121–8. [21] Kanetake N, Kume Y, Ota S, Morimoto R. Upgrading in mechanical properties of high performance aluminum alloys by compressive torsion process. Proc CIRP 2014;18:57–61. [22] Ying T, Zheng MY, Hu XS, Wu K. Recycling of AZ91 Mg alloy through consolidation of machined chips by extrusion and ECAP. Trans Nonferrous Met Soc China 2010;20:s604–7. [23] Haase M, Ben Khalifa N, Tekkaya AE, Misiolek WZ. Improving mechanical properties of chip-based aluminum extrudates by integrated extrusion and equal channel angular pressing (iECAP). Mater Sci Eng A 2012;539: 194–204. [24] Roshan MR, Mirzaei M, Jenabali Jahromi SA. Microstructural characteristics and tensile properties of nano-composite Al 2014/4 wt.% Al2 O3 produced from machining chips. J Alloys Compd 2013;569:111–7. [25] Zhang S, Xiao H, Xie H, Gu L. The preparation and property research of the stainless steel/iron scrap clad plate. J Mater Process Technol 2014;214: 1205–10. [26] Pepelnjak T, Kuzman K, Kaˇcmarˇcik I, Planˇcak M. Recycling of AlMgSi1 aluminium chips by cold compression. Metalurgija 2012;51:509–12. [27] Chiba R, Nakamura T, Kuroda M. Solid-state recycling of aluminium alloy swarf through cold profile extrusion and cold rolling. J Mater Process Technol 2011;211:1878–87. [28] Hu M, Ji Z, Chen X, Zhang Z. Effect of chip size on mechanical property and microstructure of AZ91D magnesium alloy prepared by solid state recycling. Mater Charact 2008;59:385–9. [29] Hu HE, Wang XY, Deng L. High temperature deformation behavior and optimal hot processing parameters of Al–Si eutectic alloy. Mater Sci Eng A 2013;576:45–51. [30] Sakaki S, Utsumi N. Factors causing undesirable deformations during the bending of extruded sections. Mater Trans 2006;47:1354–9. [31] Chino Y, Mabuchi M, Iwasaki H, Yamamoto A, Tsubakino H. Tensile properties and blow forming of 5083 aluminum alloy recycled by solid-state recycling. Mater Trans 2004;45:2509–15. [32] Gronostajski JZ, Kaczmar JW, Marciniak H, Matuszak A. Direct recycling of aluminium chips into extruded products. J Mater Process Technol 1997;64:149–56.
Contents lists available at ScienceDirect
Journal of Manufacturing Processes journal homepage: www.elsevier.com/locate/manpro
Technical Paper
Solid-state recycling of aluminium alloy swarf into c-channel by hot extrusion Ryoichi Chiba ∗ , Morihiro Yoshimura 1 Department of Mechanical Systems Engineering, Asahikawa National College of Technology, 2-2-1-6 Shunkodai, Asahikawa, Hokkaido 071-8142, Japan
a r t i c l e
i n f o
Article history: Received 3 October 2013 Received in revised form 1 October 2014 Accepted 15 October 2014 Keywords: Aluminium alloy C-channel Extrusion Machining swarf Mechanical property Solid-state recycling
a b s t r a c t In this study, we have investigated the possibility of the solid-state recycling of aluminium alloy machining swarf into c-channels using hot extrusion. Side milling swarf and lathe turning swarf generated from a cast Al–Si alloy ingot were cold compacted into columnar billets and successfully profile-extruded into equilateral c-channels at 600 K, under extrusion ratios of 10 and 18. The c-channels obtained at an extrusion ratio of 18 showed straight extrusion without warping, except in the front-end region. In case of the material recycled from the milling swarf at an extrusion ratio of 10, the optical microscopic study indicates the presence of coarse residual voids and cracks existing in regions where sufficient plastic strain was not introduced. In contrast, the material recycled from the same swarf at an extrusion ratio of 18 did not contain any coarse voids, rather had a density comparable to that of the original ingot. This could be attributed to the large strain of over 4.3 that was introduced in the sample recycled at the extrusion ratio of 18, as predicted by finite element analysis. Uniaxial tensile tests showed that the dense recycled material had a higher ductility than the original ingot, with a reduction of around 30% in the ultimate tensile strength. The material recycled from the turning swarf exhibited insufficient bonding among the individual pieces of the swarf, as compared to that recycled from the milling swarf under the same conditions, thereby resulting in inferior mechanical properties. © 2014 The Society of Manufacturing Engineers. Published by Elsevier Ltd. All rights reserved.
1. Introduction Currently, most aluminium scrap, including machining swarf, i.e., small chips, is remelted in furnaces and recycled into ingots or die-casting products. However, the conventional recycling method is associated with the following disadvantages: (1) the metal yield rate is very low, i.e., approximately 55% [1]. (2) Unavoidable reduction in the purity of recycled ingots at the remelting stage results in the degradation of their mechanical properties. (3) The remelting process requires significantly large amount of energy, which is unfavourable in terms of energy conservation. These limitations drive the development of alternative technologies for the recycling of aluminium scrap. Recently, solid-state recycling, which involves direct recycling of metal scrap into bulk material by using severe plastic deformation (SPD), has emerged as a potential alternative to the conventional remelting and recycling techniques [2]. The
∗ Corresponding author. Tel.: +81 166 55 8003; fax: +81 166 55 8003. E-mail address: [email protected] (R. Chiba). 1 Present address: Engineering Division, Oji Materia Co., Ltd., 20-6 Tokuda, Nayoro, Hokkaido 096-8555, Japan.
solid-state recycling method not only reduces the energy consumption but also improves the recycling efficiency dramatically (up to more than 95% for aluminium [1]). In addition, the solid-state recycling method offers the advantage of microstructural control during the recycling process, which allows the metal scrap to be recycled into materials with excellent mechanical properties. Thus far, several solid-state recycling processes have been intensively studied since 1980s, especially to recycle the machining swarf discharged from factory machine tools. A great majority of studies on this topic have focussed on recycling machining swarf into cylindrical bars [3–12] or rectangular bars [13–15], using hot/warm forward extrusion. In addition, recycling processes based on other SPD methods such as cyclic extrusion compression (CEC) [16], equal channel angular pressing (ECAP) [17–19], high-pressure torsion (HPT) [20] or compressive torsion [21], and combined use of forward extrusion and ECAP [22,23] have also been reported. Recently, another study [24] aimed at producing a composite material including nano-particles from machining chips via hot extrusion, and a new method was proposed to produce clad plates from iron chips by using hot rolling [25]. Almost all of the abovementioned methods require precompaction of the machining swarf, the technical details of which have been reported elsewhere [26]. A brief review of the relevant
http://dx.doi.org/10.1016/j.jmapro.2014.10.002 1526-6125/© 2014 The Society of Manufacturing Engineers. Published by Elsevier Ltd. All rights reserved.
2
R. Chiba, M. Yoshimura / Journal of Manufacturing Processes 17 (2015) 1–8
Table 1 Chemical composition of the AC4CH aluminium alloy used in this study [27]. Element
Cu
Si
Mg
Zn
Fe
Mn
Ti
Sb
Al
Mass%
0.01
6.9
0.37
0.02
0.13
0.01
0.13
0.001
Bal.
literature published before 2010 can be found in our previous paper [27]. Extruded sections are elongated metal products that have a complex configuration of the cross section, represented by angles and channels. Their cross-sectional shapes are non-axisymmetric and more complex than those of simple-shaped bars, e.g., cylindrical bars and rectangular bars. Extruded sections are in high demand in the industry, similar to the simple-shaped bars and metal plates. In recent years, the demand for extruded sections of aluminium alloy has been significantly increasing with the growing needs of applications demanding reduction in weight. Therefore, it is highly necessary to develop methodologies for recycling metal scrap into extruded sections. This is expected to increase the opportunities of using recycled materials, thereby leading to the promotion of use of recycled resources. However, to the best of the authors’ knowledge, there are no attempts reported so far on the direct solid-state recycling of machining swarf into extruded sections. Extrusion to form complex configuration of the cross section is different in several ways from extrusion into simple cylindrical/rectangular bars. First, if the complex configuration is asymmetric, the extrudate may exhibit warping. Second, the streamlines of the plastic flow are likely to change steeply and may bifurcate. Third, because the slit width of dies can be location dependent, there exist portions that undergo deformation for a local extrusion ratio considerably larger than the global extrusion ratio. Fourth, complex configuration generally has many corners and the contact area (friction area) of die land is very large compared to the case of simple configuration for the same extrusion ratio. In other words, a wide area undergoes significant shear deformation. Of course, the performance of the solid-state recycling is affected by the above factors. The present study investigates the possibility of recycling aluminium alloy machining swarf into c-channels using hot extrusion. Two types of machining swarfs of different shapes were coldcompacted into the shape of a billet, and subsequently respective billets were made into equilateral c-channels through hot profile extrusion process at different extrusion ratios. The surface appearance, density, microstructure, and mechanical properties of different specimens obtained under different recycling conditions were compared. Finally, we have discussed the feasibility of adopting cold-compaction followed by hot extrusion, for the solid-state recycling of aluminium alloy machining swarf into c-channels. 2. Experimental procedure
Fig. 1. Appearance of the machining swarf obtained by (a) side milling (reproduced from Fig. 1a in [27]) and (b) lathe turning.
obtained by lathe turning is found to be curled, short swarf of length 5–30 mm, width 1.5 mm, and thickness 0.5 mm. The machining swarf was ultrasonically degreased with acetone for 10 min. The cleaned machining swarf was then placed in a cylindrical container and compacted at a pressure of 303 MPa [27,28] under room temperature to form billets of diameter 20.5 mm and height approximately 25 mm. The compacts were unloaded immediately after the loading pressure reached 303 MPa. A universal testing machine (Autograph AG-X 250 kN, Shimadzu) was used for the compaction as well as for the subsequent extrusion process, described in the following section. As a pre-processing step, the compacts (billets) were annealed at 300 ◦ C for 1 h, before the extrusion process. This was performed to prevent cracks in the extrudates, which were observed in the hot extrusion of the ascompacted billets.
2.1. Machining swarf production and cold-compaction 2.2. Hot extrusion Two types of machining swarfs with different shapes were produced by side milling and turning operations of a commercial AC4CH aluminium alloy ingot with cutting oil. The chemical composition of the aluminium alloy used in this study is summarized in Table 1. In case of side milling, the cutting conditions, namely, cutting rate, feed rate, and cut depth, adopted during the machining operations were 46 m/min, 0.1 m/min, and 1 mm, respectively, while those for turning were 67 m/min, 0.1 mm/rev, and 0.5 mm, respectively. Fig. 1 shows the appearance of the machining swarf. As can be seen from the figure, the machining swarf obtained by side milling is needle-shaped and spirally twisted of length 25–30 mm and width 1 mm [27]. On the other hand, the machining swarf
In the present study, the extrusion was performed by the forward extrusion method. The dies used in this study have a C-shaped orifice at the centre (Fig. 2), so that a symmetry axis passes centrally in the dies. The dimensions x and y were kept as the same, in order to maintain the ratio between the height and width of the c-channel products as 1:1. In addition, the corner radius of 0.5 mm was set at all the eight corners. An approach region with a die angle of 45◦ was also set to get a smooth transition from the end of the container to the inlet of the dies. Two types of dies with different extrusion ratios R = 10 and 18 were prepared, which have an identical die land length of 3 mm.
R. Chiba, M. Yoshimura / Journal of Manufacturing Processes 17 (2015) 1–8
> 40 mm
3
30 mm Tab 1 mm
t
Grip section
A
A
Fig. 4. Edge-view sketch of tabbed tensile specimens.
O
y
t
E
x
x = y = 9.1, t = 1.3 and E = 4.55 for R = 10
B
x = y = 7.0, t = 1.0 and E = 3.5 for R = 18 (a) A—O—A A— O — B
Gauge section
R 0.5
Die angle 45°
t
1 (Approach) 3 (Die land)
30° Back relief angle (b) Fig. 2. Dimensions (in mm) of orifice on (a) die face and (b) cross sections along the thickness direction.
The annealed compact was placed in a cylindrical container, and then extruded into c-channels at 600 K using the universal testing machine. The temperature during the extrusion process was controlled by using a mantle heater (Tokyo Garasu Kikai Co. Ltd.) installed around the container and a desktop temperature adjuster (TC-1N, AS-ONE). Also, a MoS2 -based lubricant was applied over the inner surface of the container, and the taper and land surfaces of the dies at every extrusion, in order to prevent the increase in the extrusion load as a result of friction. The ram speed was maintained at 5 mm/min, resulting in an extrusion velocity of ∼1.5 mm/s. The compacts made of the milling swarf were extruded at R = 10 and 18, while those made of the turning swarf were extruded only at R = 18. 2.3. Testing of mechanical properties and microscopic observation Tensile specimens were cut out from the steady deformation region of the extrudates in the form of c-channels. Specifically, the extrudates were cut into side parts and a back part (Fig. 3), and each part was subjected to uniaxial tension parallel to the extrusion direction (ED), in order to evaluate the 0.2% proof stress, ultimate
tensile strength (UTS), and uniform elongation. Because the small width of the cut parts made it difficult to machine a reduced gauge section, the cut parts were tabbed prior to the testing to avoid unwanted failure in the grip sections, as shown in Fig. 4. That is, 1mm-thick AA1100 aluminium alloy tabs were bonded onto both the sides with Araldite. The longitudinal strain was measured using a strain gauge adhered to the centre of the gauge section surface. The tensile tests were conducted at room temperature at a crosshead speed of 5 mm/min on the universal testing machine. All the results reported in this study are an average of at least three tensile tests. Additionally, Vickers hardness test was performed at an applied load of 9.8 N and a holding time of 15 s using a micro-hardness tester (model-MVK, Akashi). The microstructure of the recycled materials was observed by using optical microscopy. Prior to the observation, the surfaces of the specimens were polished with emery paper and buffed with an alumina suspension. Subsequently, the specimens were etched using 0.5 ml of 40% hydrofluoric acid and 100 ml of distilled water. 2.4. Finite element analysis of extrusion process Numerical analyses of the extrusion processes were performed using a commercial finite element code DEFORM-3DTM , in order to investigate the extrusion-induced strain distribution in the transverse section (i.e., the plane perpendicular to the ED) of the extruded c-channels. Considering the symmetry of the deformation, a half of the billet was meshed with 32,000 linear tetrahedral elements. In particular, the mesh in the die inlet was maintained fine in order to resolve the large shearing of the material that takes place in this region. An isotropic rigid perfectly plastic model obeying von Mises yield criterion was used for the billet (i.e., the billet was assumed to be void-free and the density of billet was taken as the theoretical density of aluminium [15]), whereas the tools were defined to be rigid bodies. The experimental flow stress data obtained from the uniaxial compression tests of the annealed AC4CH ingot at an initial strain rate of 0.014 s−1 at 600 K were used as the deformation resistance properties in the constitutive model. The experimental stress–strain curve is presented in Fig. 5, which shows the same level of flow stress as Al–12Si eutectic alloy [29]. The friction between the billet and tools was considered to obey the Coulomb rule, with an assumed constant friction coefficient of 0.2. 3. Results and discussion 3.1. Microstructure of machining swarf and billet density
Fig. 3. Cutting of c-channel material for tensile specimens.
The microstructures of the original AC4CH ingot and milling swarf are shown in Fig. 6. The microstructure of the original ingot reveals needle-like eutectic Si phases in coarse primary Al phases. Moreover, it could be observed that the Si phases were broken into short needle-like particles of average length ∼15 m as a result of the cutting work. As described above, two types of machining swarfs were pressed to form billets of diameter 20.5 mm. The density of the billets made of milling swarf and turning swarf was found to be 2.397 g/cm3 and 2.386 g/cm3 , respectively, which correspond to relative densities of
4
R. Chiba, M. Yoshimura / Journal of Manufacturing Processes 17 (2015) 1–8
Fig. 5. Experimental and approximated stress–strain curves of annealed AC4CH ingot.
0.891 and 0.887, respectively. This indicates that the billets contain many voids. Furthermore, the macroscopic inspection of the surfaces of both the billets revealed that the size of an individual void was larger in case of billet derived from the turning swarf.
Fig. 7. Appearances of the extrudates, as seen from (a) lateral view and (b), (c) bird view.
3.2. External appearance of the recycled materials
Fig. 6. Optical micrographs of (a) original AC4CH ingot (reproduced from Fig. 9a in [27]) and (b) machining swarf obtained by side milling.
Fig. 7 shows the external appearance of the specimens obtained through the hot extrusion processes of the billets made of the milling swarf (Fig. 1a). For comparison, specimens extruded without lubrication are also shown in the figure. The surfaces of the specimens extruded with lubricant were found to be shiny and smooth for both R = 10 and 18, while those extruded without lubricant were not shiny and smooth because of the considerable friction against the dies. In some specimens extruded at R = 10, saw bladelike tears were observed at the tip of the side parts, as shown in Fig. 7a. The specimens extruded at R = 18 exhibited warping in the region less than about 30 mm away from the front end due to the end effect, but showed straight extrusion in the other region. In
R. Chiba, M. Yoshimura / Journal of Manufacturing Processes 17 (2015) 1–8
5
Fig. 8. Optical micrographs of specimens extruded from milling-swarf billets at (a), (b) R = 10; (c) R = 18; and (d), (e) extruded from a turning-swarf billet at R = 18.
contrast, the specimen extruded at R = 10 with lubrication exhibited warping throughout the length of the specimen. The surfaces of the specimens recycled from the turning swarf (Fig. 1b) had the same characteristics as those of the recycled specimens obtained from the milling swarf. Note that the authors considered only the specimens that were extruded with lubrication as a target for the following discussion. 3.3. Macro- and micro-structures of the recycled materials Fig. 8 shows the optical micrographs of the transverse section of the recycled specimens (extrudates). As can be seen from the optical images shown in Fig. 8a and b, the material recycled from the milling swarf by hot extrusion at R = 10 exhibited relatively large voids and cracks. This indicates that the bonding among the individual pieces of swarf is insufficient. On the other hand, the material recycled from the same swarf by extrusion at R = 18 had no observable cracks located at the bonded interfaces of the swarf. Therefore, the corresponding optical image revealed a highly dense surface, although minute (<5 m) voids were slightly observed. Furthermore, the optical image of the material recycled from the turning swarf by extrusion at R = 18 (Fig. 8d and e) revealed a number of voids and long cracks along the swarf interfaces in the corner parts and adjacent to the surfaces. The cracks may be attributed to the inability of the turning swarf billets to accommodate large shear
strain induced by extrusion. That is, the milling swarf, which is thinner than the turning swarf, has a higher deformability and can gain entry into tiny spaces to readily form a continuum body without voids. On the other hand, the individual voids originally included in the turning swarf billets are larger than those in the milling swarf billets, despite their comparable densities. Therefore, even an extrusion at R = 18 is not sufficient to collapse the large voids. Hence, the severe shear deformation of the material, including the residual voids near the surfaces, causes obvious cracks along the swarf interfaces (Fig. 8d and e). Fig. 9 shows the microstructure of the extruded specimens at the centre of the transverse section of the back part (see Fig. 3). As evidenced from the figure, no boundaries between the chips were found in the materials recycled from the milling swarf, whereas the boundaries were clearly observed in the material recycled from the turning swarf. It can also be seen that the eutectic Si phases were refined, when compared with those in the original ingot and machining swarf. There was no large difference in the size of Si phases between the materials recycled at different extrusion ratios (R = 10 and 18) or between the materials recycled from the differently shaped swarfs; the Si phases/particles of every recycled material ranged from 1 to 10 m in size. However, comparing Fig. 9a and b, it appears that the Si particles are more isolated in the materials recycled at R = 18, when compared with the material recycled at R = 10.
6
R. Chiba, M. Yoshimura / Journal of Manufacturing Processes 17 (2015) 1–8 Table 2 Relative density of the materials recycled from milling swarf. Billet
Relative density
0.891
Extrudates R = 10
R = 18
0.970
0.998
Fig. 10. Mechanical properties of each separated part of the recycled material in the form of c-channel.
3.4. Relative density With reference to the density of the AC4CH ingot, the relative densities of the materials recycled from the milling swarf were measured using the Archimedes method. The corresponding results shown in Table 2 imply that the specimen extruded at R = 10 still has considerable voids. This corresponds to the fact that the residual voids and cracks were observed in the specimen, as shown in Fig. 8a and b. The density of the material recycled at R = 18 was nearly identical to that of the original ingot. 3.5. Mechanical properties at room temperature
Fig. 9. Microstructure of the specimens extruded from milling-swarf billets at (a) R = 10 and (b) R = 18 (c) extruded from a turning-swarf billet at R = 18.
Fig. 10 shows the tensile test results (0.2% proof stress, UTS, and uniform elongation) of the separated parts of the respective extruded specimens. The tensile test results indicate greater variability in the uniform elongation than in the proof stress and UTS, similar to the case of solid-state recycled magnesium alloy [6]. Furthermore, major differences in the strength level and ductility between the side part and back part could be observed in all the obtained c-channels. Both the strength level and ductility were found to be higher in the side part. This could be attributed to the difference in the strain introduced during extrusion processes, as described in the forthcoming section. The 0.2% proof stress, UTS, and fracture strain of the AC4CH ingot were 133.5 MPa, 191.1 MPa, and 0.030, respectively (details not shown in Fig. 10) [27]. Thus, all the recycled materials exhibited a decrease in the mechanical strengths. However, the material recycled from the milling swarf by extrusion at R = 18 showed an increase in the ductility. The observed improvement in ductility is due to the Si-phase refinement, sufficient void reduction, and probably the partial dynamic recrystallization. Since the aluminium alloy extruded sections generally undergo secondary processing, such as bending, to form actual structural parts, their ductility is a very important factor from the application perspective [30]. The material recycled by extrusion at R = 10 had lower ductility. Some of the side parts cut out from
R. Chiba, M. Yoshimura / Journal of Manufacturing Processes 17 (2015) 1–8
7
Table 3 Vickers hardness of the materials recycled from milling swarf. Extrusion ratio
Side part
Back part
R = 10 R = 18
44.7 HV(1.0) 49.4 HV(1.0)
43.9 HV(1.0) 49.6 HV(1.0)
this recycled material fractured with little plastic deformation. This indicates that the bonding among the individual pieces of swarf is imperfect (Fig. 8a and b). However, decrease in the mechanical strength of recycled materials is not a critical disadvantage, as several applications (such as window frames) do not demand the full strength and ductility of as-cast aluminium, allowing solid-state recycled materials to be used instead [2]. Next, the effect of the shape of machining swarf on the mechanical properties of the recycled materials is discussed. As can be seen from Fig. 10, both the mechanical strength and uniform elongation of the specimen recycled from the turning swarf are inferior to that from the milling swarf. This is the synergetic effect of residual chip boundaries (shown in Fig. 9c) and a number of voids and cracks (Fig. 8d and e). The milling swarf and turning swarf differ markedly in the ratio of surface area to volume, i.e., contamination level of oxides. However, their effects on the mechanical properties of the recycled materials are minor, as the differences in the contamination of oxides have negligible effects on the mechanical strength of recycled aluminium materials at a wide range of temperature and on the ductility at room temperature [31]. As stated earlier, the main reason underlying the observed poor mechanical properties is the presence of large voids originally included in the turning swarf billets, which were not sufficiently collapsed during the extrusion process. One way to improve the bonding between the chips and reduce the residual voids and cracks in the specimen recycled from the turning swarf is to increase the extrusion temperature and decrease the extrusion rate. Gronostajski et al. [32] states that a high extrusion temperature makes the plastic flow of the material possible into pores and voids, and that a relatively low extrusion rate gives the time necessary for the diffusional transport of matter. Vickers hardness was measured at two different points on the transverse section of the specimens recycled from the milling swarf. The measurement locations are the centre of the transverse section of each separated part. The measured results are summarized in Table 3. The effect of measurement location on the hardness was very small, irrespective of the extrusion ratio. On the other hand, considering the effect of extrusion ratio, the material recycled by extrusion at R = 18 is found to be harder in both the parts. This result corresponds to the trend observed in the mechanical strength of the specimens (Fig. 10). 3.6. Numerical prediction of introduced strain Fig. 11 shows the distribution of equivalent strain in the cchannels extruded at R = 10 and 18, as predicted through the finite element analyses. For each extrudate, the contour plot is displayed in the transverse section of the steady deformation region. It can be observed that both the c-channels have strain non-uniformly distributed in the cross section and a large strain is introduced at corners. Moreover, in each extrusion case, the overall strain level is higher in the side parts than in the back part. This would lead to higher strength and ductility of the side parts in the tensile tests. A comparison of the extrudates obtained by extrusion at R = 10 and 18 presents a difference of 0.5 in the minimum value of equivalent strain. This difference improved the bonding state among individual pieces of machining swarf in the material recycled at R = 18, resulting in higher mechanical strength and elongation. Note that because the minimum equivalent (plastic) strain is given by ln R for
Fig. 11. Spatial distribution of equivalent strain in specimens extruded at R = 10 and 18. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
extrusion into simple cylindrical bars, those strain values are calculated to be 2.3 for R = 10 and 2.9 for R = 18, which are lower than the minimum values in Fig. 11 by about 1.5. As can be seen from Figs. 10 and 11, the mechanical properties of the recycled materials are positively correlated with the introduced strain. 4. Conclusions Recycled materials in the form of equilateral c-channel were obtained from the milling and turning swarf of Al–Si alloy (AC4CH) by cold-compaction followed by hot extrusion at 600 K at two different extrusion ratios of R = 10 and 18. The following conclusions can be drawn on the basis of the systematic investigation of their microstructure and mechanical properties at room temperature: (1) The materials recycled from the milling swarf via hot extrusion process at two different extrusion ratios had lower proof stress and ultimate tensile strength than those of the original ingot. However, the material recycled at an extrusion ratio of 18 exhibited an increase in the ductility because of the fine dispersion of Si particles with sufficient void reduction. (2) The mechanical strength and uniform elongation of the recycled materials were both positively correlated with the plastic strain introduced into the c-channels during the extrusion. Thus, in terms of the mechanical properties, the side parts of the channels tend to be superior to the back parts. (3) The material recycled from the turning swarf had many voids and cracks, and also chip boundaries, which resulted in inferior mechanical strength and ductility. In contrast, these defects were not included in the material recycled from the milling swarf under the same condition. This suggests that the type of machining swarf significantly affects the mechanical properties of the solid-state recycled materials. (4) The hot profile extrusion of the aluminium alloy machining swarf for an extrusion ratio of 18 or more, corresponding to an introduced equivalent strain of at least 4.3, have good potential
8
R. Chiba, M. Yoshimura / Journal of Manufacturing Processes 17 (2015) 1–8
to produce c-channels of recycled materials with comparable density and superior ductility to the original ingot. They could be potentially used in certain branches of industry, for instance, in applications such as the non-load-bearing components that do not demand the full strength of as-cast aluminium. Acknowledgement The authors would like to thank Mr. Toshiyuki Suzuki and Mr. Toshiyuki Akimoto for preparing the extrusion dies. References [1] Lazzaro G, Atzori C. Recycling of aluminium trimmings by conform process. Light Metals 1992;3:1379–84. [2] Allwood JM, Cullen JM, Cooper DR, Milford RL, Patel ACH, Carruth MA, et al. Conserving our metal energy. WellMet2050. Cambridge: University of Cambridge; 2010. [3] Wu SY, Ji ZS, Rong SF, Hu ML. Microstructure and mechanical properties of AZ31B magnesium alloy prepared by solid-state recycling process from chips. Trans Nonferrous Met Soc China 2010;20:783–8. [4] Anilchandra AR, Surappa MK. Influence of tool rake angle on the quality of pure magnesium chip-consolidated product. J Mater Process Technol 2010;210:423–8. [5] Peng T, Wang Q, Han Y, Zheng J, Guo W. Microstructure and high tensile strength of Mg–10Gd–2Y–0.5Zr alloy by solid-state recycling. Mater Sci Eng A 2010;528:715–20. [6] Zhao ZD, Chen Q, Yang L, Shu DY, Zhao ZX. Microstructure and mechanical properties of Mg–Zn–Y–Zr alloy prepared by solid state recycling. Trans Nonferrous Met Soc China 2011;21:265–71. [7] Zhang T, Ji Z, Wu S. Effect of extrusion ratio on mechanical and corrosion properties of AZ31B alloys prepared by a solid recycling process. Mater Des 2011;32:2742–8. [8] Hu ML, Ji ZS, Chen XY, Wang QD, Ding WJ. Solid-state recycling of AZ91D magnesium alloy chips. Trans Nonferrous Met Soc China 2012;22:s68–73. [9] Anilchandra AR, Surappa MK. Microstructure and damping behaviour of consolidated magnesium chips. Mater Sci Eng A 2012;542:94–103. [10] Miao J, Ye B, Wang Q, Peng T. Mechanical properties and corrosion resistance of Mg–10Gd–2Y–0.5Zr alloy by hot extrusion solid-state recycling. J Alloys Compd 2013;561:184–92. [11] Anilchandra AR, Surappa MK. Microstructure and tensile properties of consolidated magnesium chips. Mater Sci Eng A 2013;560:759–66. [12] Mindivan H, Taskin N, Kayali ES. Recycling of pure magnesium chips by cold press and hot extrusion processes. Acta Phys Pol A 2014;125:429–31. [13] Guley V, Ben Khalifa N, Tekkaya AE. Direct recycling of 1050 aluminum alloy scrap material mixed with 6060 aluminum alloy chips by hot extrusion. Int J Mater Form 2010;3:853–6. [14] Li DH, Hu ML, Wang HB, Zhao WA. Low temperature mechanical property of AZ91D magnesium alloy fabricated by solid recycling process from recycled scraps. Trans Nonferrous Met Soc China 2011;21:1234–40.
[15] Guley V, Guzel A, Jager A, Ben Khalifa N, Tekkaya AE, Misiolek WZ. Effect of die design on the welding quality during solid state recycling of AA6060 chips by hot extrusion. Mater Sci Eng A 2013;574:163–75. [16] Peng T, Wang QD, Han YK, Zheng J, Guo W. Consolidation behavior of Mg–10Gd–2Y–0.5Zr chips during solid-state recycling. J Alloys Compd 2010;503:253–9. [17] Misiolek WZ, Haase M, Ben Khalifa N, Tekkaya AE, Kleiner M. High quality extrudates from aluminum chips by new billet compaction and deformation routes. CIRP Ann – Manuf Technol 2012;61:239–42. [18] Luo P, McDonald DT, Zhu SM, Palanisamy S, Dargusch MS, Xia K. Analysis of microstructure and strengthening in pure titanium recycled from machining chips by equal channel angular pressing using electron backscatter diffraction. Mater Sci Eng A 2012;538:252–8. [19] Luo P, McDonald DT, Palanisamy S, Dargusch MS, Xia K. Ultrafine-grained pure Ti recycled by equal channel angular pressing with high strength and good ductility. J Mater Process Technol 2013;213:469–76. [20] Abd El Aal MI, Yoo Yoon E, Seop Kim H. Recycling of AlSi8Cu3 alloy chips via high pressure torsion. Mater Sci Eng A 2013;560:121–8. [21] Kanetake N, Kume Y, Ota S, Morimoto R. Upgrading in mechanical properties of high performance aluminum alloys by compressive torsion process. Proc CIRP 2014;18:57–61. [22] Ying T, Zheng MY, Hu XS, Wu K. Recycling of AZ91 Mg alloy through consolidation of machined chips by extrusion and ECAP. Trans Nonferrous Met Soc China 2010;20:s604–7. [23] Haase M, Ben Khalifa N, Tekkaya AE, Misiolek WZ. Improving mechanical properties of chip-based aluminum extrudates by integrated extrusion and equal channel angular pressing (iECAP). Mater Sci Eng A 2012;539: 194–204. [24] Roshan MR, Mirzaei M, Jenabali Jahromi SA. Microstructural characteristics and tensile properties of nano-composite Al 2014/4 wt.% Al2 O3 produced from machining chips. J Alloys Compd 2013;569:111–7. [25] Zhang S, Xiao H, Xie H, Gu L. The preparation and property research of the stainless steel/iron scrap clad plate. J Mater Process Technol 2014;214: 1205–10. [26] Pepelnjak T, Kuzman K, Kaˇcmarˇcik I, Planˇcak M. Recycling of AlMgSi1 aluminium chips by cold compression. Metalurgija 2012;51:509–12. [27] Chiba R, Nakamura T, Kuroda M. Solid-state recycling of aluminium alloy swarf through cold profile extrusion and cold rolling. J Mater Process Technol 2011;211:1878–87. [28] Hu M, Ji Z, Chen X, Zhang Z. Effect of chip size on mechanical property and microstructure of AZ91D magnesium alloy prepared by solid state recycling. Mater Charact 2008;59:385–9. [29] Hu HE, Wang XY, Deng L. High temperature deformation behavior and optimal hot processing parameters of Al–Si eutectic alloy. Mater Sci Eng A 2013;576:45–51. [30] Sakaki S, Utsumi N. Factors causing undesirable deformations during the bending of extruded sections. Mater Trans 2006;47:1354–9. [31] Chino Y, Mabuchi M, Iwasaki H, Yamamoto A, Tsubakino H. Tensile properties and blow forming of 5083 aluminum alloy recycled by solid-state recycling. Mater Trans 2004;45:2509–15. [32] Gronostajski JZ, Kaczmar JW, Marciniak H, Matuszak A. Direct recycling of aluminium chips into extruded products. J Mater Process Technol 1997;64:149–56.