This paper is subject to revision.

Statements and opinions advanced in this paper or during presentation are the authors and are his/her responsibility, not the Associations. The paper has been edited by NADCA for uniform styling and format. For permission to publish this paper in full or in part, contact NADCA, 241 Holbrook, Wheeling, Illinois, 60090, and the author.

Dross Recovery of Aluminum Alloy 380

1

Shuping Wang1, Henry Hu1, Yeou-li Chu2, and Patrick Cheng2 Department of Mechanical, Automotive & Materials Engineering University of Windsor Windsor, Ontario, Canada N9B 3P4 e-mail: huh@uwindsor.ca Department of Research and Development Ryobi Die Casting (USA), Inc. Shelbyville, IN 46176-9720 e-mail: chuy@ryobi.com, and chenp@ryobi.com
2

ABSTRACT Due to demand for cost saving and environmental concerns, recovery of aluminum dross becomes increasingly critical for die casters. However, recovery rates of the dross are often unknown to die casting shops since most dross is presently recycled externally and aluminum content in the dross depends on the practice of molten metal processing. In this study, the dross collected from both reverberatory furnace and transferring ladle was recovered with flux. The recovery rate of the recycled metal was determined based on weight measurements. To ensure the quality of the recovered aluminum, its chemistry was analyzed. The cleanliness of the recycled metal was assessed based on microstructural analysis. The tensile properties of the recovered aluminum also evaluated for the purpose of comparison. INTRODUCTION During re-melting, refining, and casting process of aluminum alloys and scraps, aluminum dross, primarily oxides and nitrides of aluminum and entrapped metallic aluminum, is generated at the surface of the molten metal resulting from its uncontrolled reaction with the furnace atmosphere at elevated temperatures [1]. Recycling of aluminium dross is one of the most challenging tasks in die casting processes since it is difficult to separate the oxides from metallic aluminum even at a high temperature. In a typical recovery process, the dross is normally melted at high temperatures in a furnace. However, at elevated temperatures, free metallic aluminum in the dross is easily susceptible to oxidation and, moreover, commonly tends to ignite and burn in the presence of air to emit toxic gases. The burning of the aluminum can decrease substantially the amount of aluminum recovered [2, 3]. The dross as a by-product not only brings huge waste, but also produces pollution to the environment. Also, due to high market demand for cost saving on die castings, the recovery of Al drosses becomes critical for die casters. However, recovery rates of the dross are often unknown to die casting shops since most dross is presently recycled externally and aluminum content in the dross depends on the practice of molten metal processing. In this study, the dross collected from both reverberatory furnace and transferring ladle in a die casting shop was recycled with flux. The recovery rate of the recycled metal was determined based on weight measurements. To ensure the quality of the recycled aluminum, the chemistry of the recovered aluminum was analyzed. The cleanliness of the recycled metal was assessed based on microstructural analysis. The tensile properties of the recovered aluminum cast in an ingot mold were evaluated.

EXPERIMENTAL PROCEDURES MELTING, REFINING AND CHEMISTRY Dross samples of aluminum alloy A380 were collected from the reverberatory furnaces and transferring ladles at Roybi Die Casting. Wedron Flux (WF132) [4] was selected and mixed with the Al dross with a certain weight ratio of dross to flux equal to 5:2. Melting of the dross and flux mixture was carried out in a 2.6 kw, 50/60 HZ electrical furnace with a maximum temperature of 1200 0C. The furnace temperature was set at 1000 0C when running the experiment to ensure the temperature of the inside crucible (melt) is kept around 900 0C. The melt was held at 900 0C for one hour, stirred for 15 minutes, and then the recovered aluminum alloy A380 was cast at 900 0C into a steel ingot mold to produce plates (608020 mm). The chemistry of cast plates was analyzed with optical emission spectroscopy. DETERMINATION OF RECOVERY RATE The dross was weighed prior to refining experiments, and the recovered aluminum alloy was weighed after the experiments. The recovery rate of the dross was determined based on the following expression: Recovery Rate (%) = (Weight of recovered Al)/ (Dross Weight) DENSITY MEASUREMENT AND MICROSTRUCTURAL ANALYSIS Following the measurement of specimen weight in the air and distilled water, the actual density (Da) of each sample with the dimensions of 10x10x10 mm was determined using the Archimedes principle based on ASTM Standard D3800 [5] Da = WaDw/( Wa-Ww) where Wa and Ww are the weight of the specimen in the air and in the water, respectively, and Dw is the density of water. The cleanliness of the recovered aluminum metal was assessed by optical metallographic observation. The specimen preparation for microstructural analysis followed the standard metallographic procedure including sectioning, mounting, grinding, polishing and etching. A Buehler optical image analyzer 2002 system was used to determine primary characteristics of the specimens. TENSILE TESTING The mechanical properties of the ingot mold-cast recovered aluminum alloy A380 were evaluated by tensile testing, which was performed at room temperature on an Instron machine equipped with a data acquisition system. Following ASMT B557 [6], subsize flat tensile specimens (25 mm in gage length, 6 mm in width, and 5 mm in thickness) were machined from the squeeze cast discs. The tensile properties, including ultimate tensile strength (UTS), 0.2% yield strength (YS), and elongation to failure (Ef) were obtained based on the average of three tests. RESULTS AND DISCUSSION CHEMICAL ANALYSIS The chemical compositions of two typical samples of the recovered metallic aluminum alloys are listed in Table 1. Sample 1 was recovered from the dross collected from the reverberatory furnaces, and Sample 3 from the transferring ladles. The results of chemical analyses show that the three key elements, Si, Cu and Fe, have a composition of about 8, 3 and 1 wt.%, respectively. The data presented in Table 1 indicate that the chemistry of the recovered metallic aluminum is compliant with the specification of aluminum alloy A380 given in reference 7.

Table 1. Chemical composition of recovered metallic aluminum compared with the specification of aluminum alloy 380
Si (wt.%) Sample 1 Sample 3 Alloy A380 [8] 8.99 8.44 7.5-9.5 Cu (wt.%) 3.27 3.23 3.0-4.0 Fe (wt.%) 1.13 1.12 1.3 max Mn (wt.%) 0.24 0.22 0.50 max Mg (wt.%) 0.03 0.01 0.30 max [9] Zn (wt.%) 1.99 1.93 3.0 max Ni (wt.%) 0.08 0.07 0.5 max Sn (wt.%) 0.02 0.02 0.35 max Al (wt.%) balance balance balance

RECOVERY RATE Table 2 lists the recovery rates of four typical Al dross samples from the reverberatory furnaces and transferring ladles. The recovery rates of the dross from the reverberatory furnaces and transferring ladle are around 55% and 83% on average, respectively. For the purpose of comparison, the data listed in Table 2 were plotted in Figure 1. It can be seen from Figure 1 that the dross from the transferring ladle has much higher recovery rates than those from the reverberatory furnaces. This observation indicated that there is much more metallic Al present in the dross collected from the transferring ladles since the dross in the ladle is generated from the aluminum melt pre-refined in the reverberatory furnaces.

Table 2. Recovery rates of recycled aluminum drosses

Sample No. 1 2 3 4 Sampling location reverberatory furnace reverberatory furnace transferring ladle transferring ladle 283 240 84.8% Dross weight (g) 390 465 394 Weight of Recovered Al (g) 213 263 321 Recovery rate: (wt. %) 54.6% 56.5% 81.5%

100

81.5
80

84.8

Recovery rate (%)

60

54.6

56.5

40

20

0
Sample 1 Sample 2 Sample 3 Sample 4

Figure 1. Recovery rate of aluminum alloy (380) drosses collected from the reverberatory furnaces (Samples 1 and 2) and transferring ladles (Samples 3 and 4).

DENSITY MEASUREMENT The density measurements of the recovered aluminum are given in Table 3. It can be seen from Table 2 that the density of the recovered aluminum is above 2.7 g/cm3 , which is slight lower than that of the die-cast A380 alloy[8]. The relatively low density is due possibly to the fact that the recovered aluminum which was cast in an ingot mould under open atmosphere may be less dense than die cast alloy 380 under applied pressure. But, it is evident that the density of the recovered aluminum is in a range comparable to that of alloy A380.

Table 3. Density measurement of recycled aluminum

Sample 1 Density (g/cm3) 2.7231 2.7332 2.7358 2.7004 2.7981 Sample 2 Sample 3 Sample 4 Alloy A380 (Die-cast)[8]

MICROSTRUCTURE Figure 2 shows the microstructures of the aluminum alloys recovered from dross collected from the reverberatory furnaces and transferring ladles. It is evident from Figure 2 that the recovered alloy is free from porosity and oxides although the microstructure such as silicon phase is relatively coarse. This observation implies that the recovered metal is clean, and could be used for casting production. To identify possible phases involved in the recovered aluminum alloy, samples 1 and 3 are further observed under the optical microscope with high magnification. The enlarged microstructures are shown in Figures 4 and 5. It is evident that the microstructure of the ingot mold-cast recovered aluminum alloy contains the primary -Al, Si phase, CuAl2, Fe containing intermetallic phases. The phase observation indicates that, despite its coarse microstructure, the recovered aluminum alloy possesses the same types of phases as those present in the die cast A380 given in

reference 8. The relatively coarse microstructure of the recovered alloy is primarily due to slow cooling taking place during ingot mold casting compared with die casting.

100 m
100 m

Figure 3. Optical micrograph showing porosity and oxide free microstructure of the aluminum alloy recovered from dross collected from (a) reverbratory furnace and (b) transfer ladles.

Al

CuAl2

Al15Fe3 Si2

CuAl2

25 m

25 m

Si
Figure 4. Micrograph showing different phases present in the aluminum alloy recovered from dross collected from the reverbratory furnaces.

CuAl2

(Al5Fe Si)

(Al15Fe3 Si2)

Si Al
25 m 25 m

Figure 5. Micrograph showing different phases present in the aluminum alloy recovered from dross collected from the transferring ladles.

TENSILE PROPERTIES Representative averaged tensile properties, UTS, YS and Ef , of the recovered aluminum and the die cast alloy A380 are given in Table 4. Figure 2 shows the tensile properties of the recovered aluminum compared to those of the die cast alloy A380 at room temperature. It can be seen that the UTS, YS and Ef , of the recovered aluminum are around 170 MPa, 120 MPa and 1% on average, which are slightly slower than those of the die cast A380. However, the tensile properties for both the recovered aluminum and the die cast A380 are at a comparable level. The coarse microstructure should be responsible for the relatively low properties of the recovered aluminum alloys.
Table 4. Averaged tensile properties of the recovered alloy and the alloy A380

Sample 1 UTS (MPa) YS (MPa) Ef (%) 174.12 129.25 0.96

Sample 2 160.55 117.35 0.86

Sample 3 171.32 116.13 1.21

Alloy A380 (Die-cast)[8] 182.18 136.02 1.11

240 220 200 180 160 UTS YS Ef

2 1.8 1.6 1.4 1.2 1

Strength (MPa)

140 0.8 120 100 80 60 0.6 0.4 0.2 0

Sample 1

Sample 2

Sample 3

A380

Figure 2. Tensile properties of the recovered aluminum alloy in comparison with Alloy A380.

CONCLUSIONS From this study, it can be concluded that - the chemistry of the recovered metallic aluminum from dross collected from both the reverberatory furnaces and transferring ladles both is compliant with the specification of aluminum alloy A380; - the dross from the transferring ladle has much higher recovery rates (83%) than those (55%) from the reverberatory furnaces; - the recovered alloy is clean and free from porosity and oxides although its microstructure is relatively coarse compared to that of the die cast A380; - the microstructure of the recovered aluminum alloy contains the primary -Al, Si phase, CuAl2, Fe containing intermetallic phases, which are almost the same as those present in the die cast A380; and

Elongation (%)

the density and tensile properties for both the recovered aluminum and the die cast A380 are at a comparable level.

REFERENCES

1. H. Y. Ghorab, M. Rizk, A. Matter, and A. A. Salama, Characterization and Recycling of Aluminum Slag,
Polymer-Plastics Technology and Engineering, Vol. 43, No. 6, 2004, 1663-1673.

2. J.Y. Hwang, X. Huang, and Z. Xu, Recovery of Metals from Aluminum Dross and Saltcake, Journal of
Minerals & Materials Characterization & Engineering, Vol. 5, No. 1, 2006, 47-62,

3. B. Lucheva, Ts. Tsonev, R. Petkov, Non-Waste Aluminum Dross Recycling, Journal of Chemical
Technology and Metallurgy, Vol. 40, No. 4, 2005, 335-338

4. Material Safety Data Sheet, WF132, Wedron Flux, Black Lab Corporation, 2005. 5. Standard Test Method for Dry and Wet Bulk Density, Water Absorption, and Apparent Porosity of Thin 6. 7. 8. 9.
Sections of Glass-Fiber Reinforced Concrete, C948-81, ASTM Standards, ASTM, Vol 04.05, 2002, pp. 588-589 Standard Test Methods of Tension Testing Wrought and Cast Aluminum- and Magnesium-alloy products, B557M, ASTM Standards, ASTM, Vol 02.02, 2002, 424. Aluminum and Aluminum Alloys, ASM Specialty Handbook, ASM International, 2002, 722 H. Hu, Y. Wang, Y. Chu, P. Cheng, and A. T. Alpas, Solution Heat Treatment of Vacuum High Pressure Die Cast Aluminum Alloy A380, NADCA Transactions, 2005, 22-33. www.diecasting.org/releases/380.htm.