GRAIN REFINEMENT OF PERMANENT MOLD CAST COPPER BASE ALLOYS

Grain refinement behavior of copper alloys cast in permanent molds was investigated. This is one of the least studied subjects in copper alloy castings. Grain refinement is not widely practiced for leaded copper alloys cast in sand molds. Aluminum bronzes and high strength yellow brasses, cast in sand and permanent molds, were usually fine grained due to the presence of more than 2% iron. Grain refinement of the most common permanent mold casting alloys, leaded yellow brass and its lead-free replacement EnviroBrass III, is not universally accepted due to the perceived problem of hard spots in finished castings and for the same reason these alloys contain very low amounts of iron. The yellow brasses and Cu-Si alloys are gaining popularity in North America due to their low lead content and amenability for permanent mold casting. These alloys are prone to hot tearing in permanent mold casting. Grain refinement is one of the solutions for reducing this problem. However, to use this technique it is necessary to understand the mechanism of grain refinement and other issues involved in the process. The following issues were studied during this three year project funded by the US Department of Energy and the copper casting industry: (1) Effect of alloying additions on the grain size of Cu-Zn alloys and their interaction with grain refiners; (2) Effect of two grain refining elements, boron and zirconium, on the grain size of four copper alloys, yellow brass, EnviroBrass II, silicon brass and silicon bronze and the duration of their effect (fading); (3) Prediction of grain refinement using cooling curve analysis and use of this method as an on-line quality control tool; (4) Hard spot formation in yellow brass and EnviroBrass due to grain refinement; (5) Corrosion resistance of the grain refined alloys; (6) Transfer the technology to permanent mold casting foundries; It was found that alloying elements such as tin and zinc do not change the grain size of Cu-Zn alloys. Aluminum promoted b phase formation and modified the grain structure from dendritic to equiaxed. Lead or bismuth reduces the size of grains, but not change the morphology of the structure in Cu-Zn alloys. The grain size of the Cu-Zn-alloy can be reduced from 3000 mm to 300 mm after the addition of aluminum and lead. Similar effects were observed in EnviroBrass III after the addition of aluminum and bismuth. Boron refined the structure of yellow brasses in the presence of iron. At least 50 ppm of iron and 3 ppm of boron are necessary to cause grain refinement in these alloys. Precipitation of iron from the melt is identified as the cause of grain refinement. Boron initiates the precipitation of iron which could not be explained at this time. On the other hand zirconium causes some reduction in grain size in all four alloys investigated. The critical limit for the zirconium was found to be around 100 ppm below which not much refinement could be observed. The mechanism of grain refinement in the presence of zirconium could not be explained. Grain refinement by boron and iron can remain over a long period of time, at least for 72 hours of holding or after remelting few times. It is necessary to have the iron and boron contents above the critical limits mentioned earlier. On the other hand, refinement by zirconium is lost quite rapidly, some times within one hour of holding, mostly due to the loss of zirconium, most probably by oxidation, from the melt. In all the cases it is possible to revive the refinement by adding more of the appropriate refining element. Cooling curve analysis (thermal analysis) can be used successfully to predict the grain refinement in yellow brasses. The precipitation of iron in the liquid metal causes the metal to solidify without undercooling. Absence of this reaction, as indicated by the time-temperature (t-T) and its first derivative (dt/dT) curves, proved to be an indicator of refinement. The viability of the technique as an on-line quality control tool was proved in two foundries. The method can also correctly predict the onset of fading. The corrosion resistance of the grain refined alloys was measured in two solutions having different hydrogen activities, pH 6 and pH8, and compared with the base alloys. Potentiodynamic polarization and long term weight loss experiments were conducted to evaluate the corrosion resistance. Cu-Zn alloys were evaluated for dezincification. In general, the grain refined alloys performed marginally better than the base alloys