THE EFFECTS OF MG, CU AND CA ON THE ROLLING AND DRAWING PERFORMANCE OF 5XXX ALUMINUM ALLOY WIRE
Ye Wujun Victor
Indalco Alloys, A Division of Lincoln Electric of Canada LP[1]
ABSTRACT. In the rolling and drawing procession of 5xxx series aluminum alloy wire, chevron and breaking are problems more frequently met. Several failure cases were analyzed, alkali element Ca worsens continuous casting and rolling processions, Ca rich phase solidifies on the level control part and cause problems. Mg micro-scale segregation causes local site hardened and form chevrons on the surface of the wire, track amounts of Cu causes wire brittle during drawing. Mg, Ca and Cu segregation is caused by that not uniformly cooling during continuous casting.
Key words: Aluminum, continuous casting, rolling, failure analysis
1. Introduction
Alkali elements, Li, Na and Ca even in very low level, for example 2-5ppm, could greatly affect the hot rolling performance of 5xxx series aluminum alloys. Sodium and calcium caused the segregation to grain boundaries [1,2,3]. For sodium, even at 0.6ppm it caused 200°C to 400°C plastic property drop a lot and caused chevron which worsen the draw-ability of the product, for example, rod [4]. The impurity elements also could cause casting problem for continuous casting and rolling processions. Several processions, such as fluxing, have been developed for removing the impurity elements. In practice, it gives some way to get more pure 5xxx series alloy product. For continuous casting and rolling re-drawable rod, there are still some another factors which cause some problem, for example, trace element segregation, lubrication and internal stress. This paper is based on several cases of failure during casting, rolling and cold drawing of 5xxx series aluminum alloy rod wire to investigate the element and their distribution effects on the casting, rolling and drawing performances.
2. Experiments and failure analysis
2.1 Effect of Ca on the continuous casting procedures
For the continuous casting of aluminum alloy bar section area is about 50cm2, the even level of molten metal in the tundish is required. The precipitation of some solid phases on the tube surface through which the molten metal flow over often cause the problem for the level control, for example, the continuous bar broken or chevron and other defect on the final rod product. For the failure case of the level control, some samples were taken from the surface precipitation stuff of the tube and tundish and were analyzed with scanning electronic microscope (SEM).
2 - 1
Fig 1. The precipitation stuff EDX analysis, the element count in the bright area. Fig 2. The precipitation stuff EDX analysis in the more dark area
2 - 1
The analysis results show that the coatings on the surface of tundish are made of Al-Mg and some other elements, including Ca-Ti-Cr-V. The coatings on the surface of collie hat are made of Al-Mg and, including Ca-Ti-Cr-V, but the concentrations of Ca and Ti of the surface precipitation stuff on the collie hat is more than that in the coating on the surface of tundish, fig 1 and 2. Ti and Cr are the elements for 5xxx series alloy and Ca and V are impurity elements for that kind of alloy. With the high melt point for all other three elements, Ca in fact promotes the process of solidification of the coatings. Because in the molten metal, Ca concentration is only 0.0009%, the concentration is not high enough for the Ca rich phase to precipitate according to the phase diagram of Al-Ca, Al-Cr, Al-Ti and Al-V. But because the materials of the refractory wall content some CaO, CaCO3, there may be the reactions happened between Ca in molten metal and refractory wall.
Ca++ + 2e Ca
The high concentration of Ca caused the high trend of solidification of Ca rich phases on the wall. Because of the coating with the precipitation of Ca, Ti, Cr and V rich phases on the surface of collie hap and tundish, the inner diameter of the tube was reduced and the level of the molten metal in the tundish was not even enough to get a good solidification. This usually causes bar broken. And also, Ti and V are all the elements to be easily frozen to the wall of tundish and collie hat and form the frozen cores in the procession of solidification.
With low Ca level molten 5xxx series alloy, the trend to precipitate on the wall of the tundish and collie hat is much low than that with some little bit high Ca level. When the Ca concentration was controlled, the problems were then fixed.
2.2 The root cause of the surface scale generated during cold drawing of 5xxx series aluminum alloy wire
2.2.1 Observations
Surface scale is one type of failures during the cold drawing of aluminum alloy wire, several cases were analyzed on 5xxx, 6xxx and 2xxx series alloys. Most of them were connected with the bad distribution of the hardening elements such as Mg, Cu, even with the chilly cold continuous cast and hot rolling rod.
3 pieces of 6.40mm dia. 5356 wire samples were analyzed for the root cause of the scales on the wire surface generated during drawing. Samples were cut from 3 pieces of 6.40mm wire for SEM and microscope observation. The samples for micro-structure observation were cross sectioned in two directions, approximately vertical each other. The microscope observations on cross sections show that there are no inclusions and porosity were found, except the edge area. SEM observations on cross sections also show that no obvious micro segregation was found. The appearance of the scales is shown in fig 3. On the surface of the sample, many small fissures were found. The problems are that during the drawing process, there were some surface cracks generated and formed scales on the surface.
The analysis on the surface shows that many small cracks initiated at the site where Mg concentration is more than the average and the color is more dark than the matrix in the observation in the style of back scatter electronic image. The second hard phase particles are intermetallic which are crystal lattice match or semi-crystal lattice match to the matrix phase. The boundary between the second hard phases and matrix usually the week area, during cold drawing, there was not enough plastic for the coordinate deformation and therefore formed the fissures.
As shown in fig 4 and 5, some cracks just located on the boundary between second phases and matrix and some big cracks are not connected to the segregation of second phases, but they located just between the prolonged second phase segregation stripes. Because from the samples some of the surface cracks are between the stripes and not along the boundary between the second phases and matrix, that means at least there are over one mechanism the fissures initiated with.
Compare photographs of element distribution, the contour of Mg distribution is partially matching the contour of crack. This is indirect proof that some of the cracks between the prolonged Mg rich second phase strips initiated at the boundary of Mg rich phases and matrix.
The width of the crack in the direction approximately vertical to the longitudinal direction of the wire is about 360 um, and the distance between the adjoined second phase segregation stripes is about 590um.
Fig 3. Appearance of the scales on 6.40mm wire. Fig 4. Small cracks initiated at the site where Mg is rich.
Fig 5. 100X The cracks initiated at the site where Mg is rich and Mg content in half qualitative analysis is about 12%, more than the nominal composition -- Mg 5%.
2.2.2 Discussion on the fissure initiation on the wire for 5xxx series aluminum alloy
The microstructure of continuous casting and rolling rod is the mixture of the dynamic re-crystallized and deformed f.c.c structure. At the end of the rolling, the temperature of the rod is about 380°C, just a little bit above the re-crystallization point. Therefore during rolling all the Mg segregation and grains would be prolonged and the dynamic re-crystallization partially happened. So, the Mg segregation phase would cause some longitudinal defect but not vertical to the direction in macro scale.
In the following cold drawing procession, there are two kinds of texture for cold drawing aluminum alloy wire, internal and surface textures formed, and the intermetallics phase in 5356 aluminum alloy, the major part of them are Al3Mg2 which is f.c.c. lattice structure as well, but the lattice distance is big than that of matrix of 5356 alloy. During the rotation of the crystal, it is easy cause some internal stress along the bound between intermetallics phases and matrix. As shown in fig 3,4. There are many micro scale fissures, but there are macro-scale fissures that generated between the prolonged Mg rich phases. The reason for that is the prolonged Mg rich phased stopped coordinated deformation of the surface of the wire, some inclusions and Mg rich isolated phases is the source site of the macro scale fissures. So, if we could decrease Mg segregation for the cast bar, we could decrease the segregation of rod therefore to decrease the trend of surface fissures.
2.3 The effect of residue copper on the draw-ability of 5xxx series alloy wire
Several cases of failure analysis were carried on 5052 alloy rod. The breakage were collected from the fabrication of 0.64mmX9.5mm strip, cold drawing and rolling from 5052 9.5mm rod. Some similar phenomena happened during the wire drawing 5356 alloy as well. Other than the root cause of chevron and inclusion cluster defects, there are some other factors to cause the breaking. The statistics data shown that if the alloy made with the furnace some residue copper was there, the frequency of wire breaking is high. But if the alloy made with the furnace some residue Si there, the frequency of wire breaking is low.
Copper, 0.15 to 0.40% wt, in 6061 aluminum alloy is as an aging strengthening element. Some research results [5] show that Cu has the hardening function for that kind of alloys even in a low concentration. For 5xxx series alloy, there are always some residue elements left in the alloy, for example Si, Fe and Cu. In micro scale, there were some trends to form hard phases. The breakage analysis shown that there are some fissures on the surface of the 5xxx series alloy wire and the breaking usually happened in the hardening element rich site.
Fig 6. The fissures on the surface of 5052 wire. Fig 7. The surface appearance of the fracture surface.
Fig 8. Back scatter electronic photo of the fissures on the surface of 5052 wire.
The fracture surface of the wire is in the shear style. There are some slide bands along the longitudinal direction. The micro cracks are always in the site where there are some Ca, Cr and Cu island-like phases. That Cr and Cu segregated in micro scale and formed some hard phases which peeled off caused some fissures during cold drawing.
3. Conclusions:
For continuous casting and rolling processions, calcium promote the precipitation of the hardener element to the surface of the level control part if the refractory wall contents calcium, decrease calcium level in the molten metal could get better level control. Magnesium macro scale and micro scale segregation could cause surface fissure and worsen the surface texture structure which could cause some cracks and scales during the cold drawing process for 5xxx series aluminum wire. Residue copper in a very low level can cause the breaking during cold drawing.
References
[1] K.Horikawa, S.Kuramoto and M.Kanno, Acta Mater., 49, 3981-3989 (2001).
[2] Guang-Hong Lu, Masanori Kohyama and Ryoichi Yamamoto, Mat. Res.Soc. Symp. Proc., 653, Z7.17.1-Z7.17.6(2001)
[3] Zi-kui Liu, Effect of impurities on the processing of aluminum alloys, Industrial Technologies Program, US Department of Energy Washington, D.C. 20585
[4] Research results in Kannno Lab., Japan, 2003
[5] M.Murayama, K.Hono, W.F. Miao, and D.E. laughlin, Met Trans, Volume 32A, 2001,pp239-246
2 - 1
Indalco Alloys, A Division of Lincoln Electric of Canada LP
929 Gana Court, Mississauga, Ontario, Canada l5S 1N9
Voice (905)564-1151 ext 115
Fax (905) 564-1405
E-mail: