Antimony in aluminum scrap.

Since 1978, some European foundries have used antimony (Sb) to produce a permanently refined Al-Si eutectic alloy for the production of certain automotive castings, including wheels, brake master cylinders, connecting rods, suspension arms and brake calipers.

However, U.S. foundrymen have expressed concern that trace amounts of Sb could show up in secondary aluminum ingot and affect the mechanical properties of castings.

Although the secondary aluminum industry segregates scrap to minimize arbitrary mixing of different scrap types, some mixing does occur. Trace elements from the chemistry of one type of aluminum can deleteriously affect the properties of another type of aluminum. Trace elements may originate from a variety of sources, including the primary ore; master alloys used to alloy, grain refine or modify certain alloys; fluxes used in treating metal; and the recycling stream.

The approach to controlling impurities in a given alloy consists of: * removing the impurity, if possible for

example, magnesium levels are

typically lowered by chlorine to acceptable

concentration levels in casting

alloys; * diluting the impurity by adding impurity-free

scrap or pure primary aluminum; * adding another agent to offset the

effects of the impurity-strontium (Sr)

may be added to certain types of

alloys, with the amount added being

dependent on the concentration of

phosphorus (P) in the alloy. Research Program

As a result of this concern over Sb presence, a research effort was undertaken by Argonne in collaboration with the American Foundrymen's Society to determine the effect of trace amounts of Sb on the mechanical properties of Sr modified A-356 ingot.

The overall objective of this research was to determine whether the presence of Sb in the recycle stream could pose a materials supply problem, and, if so, to determine a least-cost method for handling the potential problem to provide for continued or expanded recycling of aluminum. The experimental portion of this research was performed in the laboratories of the AFS Cast Metals Institute. interactive Effects of Sb and Sr

Previous research addressed the interactive effects of Sr and Sb on the microstructure of laboratory alloys. However, the effects on the mechanical properties and castability of commercial alloys had not been addressed.

Controlled experiments were conducted to determine the impact that trace amounts of Sb might have on the mechanical properties (elongation, yield strength and ultimate tensile strength) and the fluidity of Sr-modified commercial A-356 ingot.

The experiments were conducted in two phases under the direction of Lee Tuttle (GMI Engineering & Management Institute) at the AFS/CMI laboratory. in the first phase, Sr modification of Sb-bearing A-356 alloy was attempted at five target levels of Sb concentration (0%, 0.05%, 0.10%, 0.15% and 0.20%). Figure 1 demonstrates that when the alloy is free from Sb, Sr modification results in an elongation of about 6-7%.

At Sb concentrations of around 400 ppm, the presence of Sb interfered with the modification that would have otherwise been expected with the addition of Sr. Trace amounts of Sb did not affect the ultimate tensile strength or the yield strength of the Sr-modified A-356. The fluidity was also affected by the amount of Sb present, decreasing at Sb concentrations of about 500 ppm.

A second set of experiments was conducted to determine the effects of trace amounts of Sb present in the range of 0-500 ppm. The effect on elongation was similar to the results from the first set of experiments; however, the minimum elongation in the second set of experiments occurred at a lower Sb concentration (Fig. 2). The reason for this was believed to be attributable to two factors: * the amount of Sr retained in the metal-an

average of 50 ppm for Phase 2 and

an average of 100 ppm for Phase 1; * the amount of residual P-an average

of about 10 ppm for Phase 2 and

about 20 ppm for Phase 1.

Previous research has shown that one result of the interaction of Sb and Sr is the formation of an intermetallic Mg[sub.2]Sb[sub.2]Sr. This was demonstrated by normalizing the data in terms of excess Sb and excess Sr where zero (0) is consistent with the formation of the intermetallic Mg[sub.s]Sb[sub.2]Sr with all available Sr and Sb in the melt. Generally, there appears to be a coincidence of the minimums in elongation of the Phase 1 and Phase 2 experiments (Fig. 3) although the Phase 1 minimum lies below that of Phase 2.

In both Phases 1 and 2, as the Sb concentration increased, the intermetallic Mg[sub.2]Sb[sub.2]Sr was expected to form in greater amounts. This may have contributed to the degradation in elongation by three mechanisms: loss of modification by consumption of the Sr, depletion of the Mg and formation of intermetallic inclusions.

In addition, as more of the Sr was bound as part of the intermetallic, there was less free Sr in the melt to counteract the residual P content. Once the maximum amount of intermetallic was formed, elongation recovered from an excess of Sb. In Phase 2, less intermetallic was formed at each concentration of Sb because less Sr was available. There also was less residual P in the Phase 2 metal.

A question that remains to be resolved is: If the intermetallic is formed, does the presence of the intermetallic in itself cause a degradation in elongation, or does the formation of the intermetallic simply reduce the amount of Sr that would otherwise be required to achieve modification of the alloy?

Although additional experimentation would be required to answer this, it appears that formation of the intermetallic served to deplete the Sr rather than to contribute to the degradation in elongation within the range of Sr, Sb and P contents of these experiments.

Previous work indicated that the amount of Sr required to achieve modification is defined by the following equation (all terms are in ppm):

Sr = [similar to] [20 + (3*P)] (1)

This equation closely fits the results of the Phase 1 and 2 experiments with a slight modification to account for the formation of the intermetallic:

Sr = [similar to] [20 + (3*P) + (Sb/2.8)] (2)

This equation implies that if Sb is present in the metal, modification of the metal with Sr can be achieved if enough Sr is added in addition to the amount of Sr that will be bound stoichiometrically to the Sb. While this is consistent with the experiments conducted, proof of this equation would require experiments to determine where the elongation recovered after it had been degraded.

Although the presence of Sb in reclaimed aluminum was anticipated, the evidence to date has been largely anecdotal. A testing program was recently initiated by the Aluminum Recycling Assn. in cooperation with Argonne to track the presence of Sb in the aluminum recycle stream.

Conclusions that we drew from these recent experiments were: * Sb concentrations as low as 100 ppm

can significantly affect the elongation

of Sr-modified A-356 alloy; * the formation of the intermetallic

Mg[sub.2]Sb[sub.2]Sr appears to prevent modification

but does not necessarily exacerbate

the degradation in elongation; * the addition of excess Sr consistent

with the P content of the metal is

expected to result in modification in

the presence of the intermetallic within

the range of Sr, Sb and P concentrations

of these experiments. Conclusions

Further experimentation by Argonne National Laboratory at AFS/CMI facilities will be designed to investigate the validity of Equation 2. One aspect of the program that has been preliminarily investigated was the lowering or elimination of antimony from reclaimed scrap aluminum. The research conducted indicated that Sb in concentration as low as 100 ppm would degrade the elongation of Sr-modified A-356.

However, the effect on mechanical properties can be offset by the addition of excess Sr. Antimony, which is referred to as a "permanent" refining attitude, has been thought to be unremovable from aluminum. However, World War II era literature uncovered by Argonne indicated that calcium metal is an efficient scavenger for antimony (and probably other tramp metals) from molten aluminum.

Recent work at the AFS/CMI labs showed that molten aluminum, containing as much as 2800 ppm Sb, could be reduced to the range of 50 ppm Sb when treated with as little as 1500 ppm of Ca. Apparently, the Ca and Sb form an insoluble intermetallic compound that separates out. This intermetallic probably combines with the aluminum oxide surface dross.

Although these results are preliminary, they do point toward significantly reducing antimony contamination from molten aluminum. A paper that details these research findings will be presented at the 1991 AFS Casting Congress in Birmingham, Alabama.