Mg-Al-Mn- (Zn) Ferric chloride Marked Nucleation by Fe-Al- Mn compounds and carbon Zr, Be Requires manganese Ferric chloride, iron-zinc alloy Very marked Nucleation by iron compounds Al, Si, Th Iron-zinc requires manganese content of 1% Mg-Zn(-Re- Mn) Ammonia Very marked Nucleation by hydrogen Al, Si, Th Gassy metal; Manganese can exceed 1%. Calcium + nitrogen Mild . . . . . . . . . Mg-Mn Zirconium Increases with decreasing manganese Nucleation by zirconium or Zr- enriched Mg Al, Si, Fe, H, Sn, Sb, Co, Ni, Mn, others . . . Source: Ref 104 (a) Elements in parentheses interfere to a lesser extent. Reference cited in this section 104. E.F. Emley, Principles of Magnesium Technology, Pergamon Press, 1966 Grain Refining of Copper Alloys In general, the grain refinement of copper alloys is not practiced as a specific molten metal processing step per se, because a certain degree of refinement can be achieved through normal casting processes. As with aluminum alloys, grain refinement in copper alloys can be achieved by rapid cooling, mechanical vibration, or the addition of nucleating or grain growth restricting agents. Further, many commercial copper alloys have sufficient solute (zinc, aluminum, iron, tin) to achieve constitutional supercooling during solidification. In this case, grain nucleation and growth are naturally retarded. Commercially pure copper can be grain refined by small additions (as little as 0.10%) of lithium, bismuth, lead, or iron, which provide constitutional supercooling effects (Ref 105, 106). Copper-zinc single-phase alloys can be grain refined by additions of iron or by zirconium and boron (Ref 107). In the latter case, the probable mechanism is the formation of zirconium boride particle nuclei for grain formation. In one case, the vibration of a Cu-32Zn-2Pb-1Sn alloy improved yield and tensile strengths by about 15%, with a 10% reduction in grain size from the unvibrated state. In general, the α copper-zinc alloys (<35% Zn) exhibit grain size reduction and greater improvement in properties, while the α-β alloys do not. Copper-aluminum alloys have been effectively grain refined with additions of 0.02 to 0.05% B; the effective nucleating agent is boron carbide (B 4 C). Figure 71 illustrates the improvement in mechanical properties achieved by grain refining a Cu-10Al alloy. Fig. 71 Effect of boron-refined grain size on the mechanical properties of Cu- 10Al alloy. Test specimens were removed from the center or the top of the ingot as indicated. Source: Ref 107. Tin-bronze alloys have been successfully grain refined by the addition of zirconium (0.02%) and boron master alloys (Ref 108). However, pressure tightness is reduced because in these long freezing range alloys, finer grain size concentrates porosity because of gas entrapment. References cited in this section 105. A. Cibula, Grain Refining Additions for Cast Copper Alloys, J. Inst. Met., Vol 82, 1953, p 513 106. G.C. Gould, G.W. Form, and J.F. Wallace, Grain Refinement of Copper, Trans. AFS, 1960, p 258 107. R.J. Kissling and J.F. Wallace, Grain Refinement of Copper Alloy Castings, Foundry, June-July, 1963 108. A. Couture and J.O. Edwards, Grain Refinement of Sand Cast Bronzes and Its Influence on Their Properties, Trans. AFS, 1973, p 453 Modification of Aluminum-Silicon Alloys In the aluminum foundry alloy family, the aluminum-silicon alloys form the bulk of the commercially important applications. Silicon is the principal alloying element in these materials. For the purposes of this discussion, these alloys will be grouped as either hypoeutectic (~7% Si) or hypereutectic ( ≥ 12% Si). A two-phase structure exists for both classes. The hypoeutectic alloys consist of a primary -aluminum matrix plus a second phase based on the silicon eutectic. In the hypereutectic alloys, silicon is the primary phase (Fig. 72). In either case, it is the morphology of the silicon phase that gives rise to the practice of modification. The silicon phase in hypoeutectic alloys, which is the eutectic constituent, solidifies as an acicular or platelike structure. In the hypereutectic alloys, the primary silicon phase is very blocky. These morphologies are detrimental to most aluminum mechanical properties, causing embrittlement, and also result in poor machinability. In addition, unmodified silicon structures are associated with increased porosity in the raw casting. Fig. 72 Aluminum- silicon binary phase diagram showing common foundry and die casting alloy compositions 319, 356, 380, 390, and the eutectic alloy 413. Source: Aluminum Smelters and Refiners Inc. Therefore, modifiers are deliberately added to the molten metal to change the shape of the silicon phase (eutectic in hypoeutectic alloys, primary silicon in hypereutectic) to a more well-rounded shape that is also more coherent with the matrix. This change also disperses shrinkage during solidification and greatly improves mechanical properties and machinability. Element additives such as sodium, calcium, strontium, and antimony are currently used to modify the silicon phase in hypoeutectic alloys, and phosphorus is used to refine the silicon phase in hypereutectic alloys. Theory of Modification The work of Crossley and Mondolfo (Ref 109) has been instrumental in describing how different impurities nucleate different solid phases. When sodium is present, the nucleation of the eutectic silicon takes place at approximately 3 to 12 °C (5 to 20 °F) undercooling (a factor that thermal analysis takes into account; see the section "Thermal Analysis" in this article). The specific degree of supercooling indicative of complete modification is different for each alloy. For 356 alloy, a typical modified eutectic temperature is about 568 °C (1054 °F). In hypoeutectic alloys, an AlSiNa precipitate appears to serve as the nucleation substrate for the silicon eutectic phase (Ref 110, 111). An AlP precipitate seems to be present even in hypoeutectic alloys (due to residual phosphorus). Sodium neutralizes AlP so that easy nucleation of silicon is prevented. In the hypoeutectic alloys, another contributing mechanism is thought to be the influence of the modifier element (for example, sodium) on restricting the growth of nuclei by reducing liquid surface tension within the eutectic phase which consists of a lamellar structure of both aluminum and silicon (Ref 6). The reduction in surface tension increases the contact angle between the aluminum and the silicon, permitting the aluminum to wet the silicon phase overall more easily and therefore restrict favored growth along one particular axis. Sodium apparently reduces the rate of diffusion of silicon in the liquid, thus restricting nuclei and grain growth. Hypereutectic Alloys. Silicon phase modification in these alloys by phosphorus, which is referred to as refinement, is theoretically a nucleation phenomenon that influences the size and shape of the primary silicon only. Phosphorus reportedly combines with aluminum to form submicroscopic AlP precipitate whose crystallographic characteristics are similar to those for silicon and that permit the primary silicon crystal to nucleate more easily and grow evenly during solidification. It is also apparent that metal casting temperature, mold temperature, overall metal composition, and turbulence of the flowing metal all play a role in determining the extent of modification (Ref 6, 112, 113). Thermal Analysis Thermal analysis is used to predict or analyze the anticipated degree of modification to be expected from a melt (Ref 114). The principles are based on the thermal arrests and degree of undercooling that take place when a melt sample is allowed to solidify under carefully controlled conditions. A thermocouple is placed in a precise position within a mold containing a melt sample, and it monitors the temperature as a function of time as the sample solidifies. A characteristic curve can be obtained, as illustrated in Fig. 73 for an unmodified alloy. The break in the curve occurs at the eutectic temperature T e when a two-phase liquid plus solid mixture X or B phase solidifies to α plus β. The beta is the eutectic silicon. In the presence of a modifier, the eutectic temperature is depressed, as shown in Fig. 74, and this depression can be correlated with the amount of modification that can be expected to take place in the full casting if the same solidification conditions are met. Fig. 73 Binary phase diagram (a) and thermal analysis curve (b) for an alloy having complete solubility in the liquid state and incomplete solubility in the solid state. Source: Ref 112. Fig. 74 Characteristic thermal analysis curve showing temperature depression that occurs after modification. Source: Ref 114. Other features of thermal analysis that can be monitored are the times associated with each thermal change. The eutectic temperature depression is influenced by alloy content, with T e = 1.1 (Mg) + 0.25 (Cu) + 0.18 (Fe) (Ref 114). The temperature excursions and the times associated with the thermal features can be related to grain size and modification (Ref 115). The thermal arrest T L associated with the liquidus curve (Fig. 73) can be correlated with grain size. The longer the undercooling time, the lower the nucleation potential and the larger the grain size. Figure 75 indicates the correlation of grain size with several grain-refiner additions as obtained with the one thermal analysis instrument. Fig. 75 Schematic showing aluminum grain size (ASTM E 112) as eval uated and displayed by one thermal analysis instrument. Source: Ref 116. Two principal instruments are commercially available for thermal analysis. Both consist of a sample cup, thermocouple, and microprocessor unit. Regardless of the instrument used, it is important that the test be conducted consistently each time for accurate results (Ref 114, 115, 116, 117). Alternate Techniques to Determine Modification There are other nondestructive techniques capable of assessing the degree of modification in a solidified casting or in regions of a casting. Such an analysis can serve as a useful foundry check. Electrical resistivity or conductivity has been found to correlate well with degree of modification (Ref 118 and 119). Conductivity has been found to increase linearly with degree of modification in Alloys A444 and A356, but detection capability decreases as the magnesium level increases. Surface finish is also a major factor. Polished surfaces display greater sensitivity and response to this technique than as-cast or even machined surfaces. Ultrasonic measurement has also been investigated, but thus far the technique has not been found to provide adequate sensitivity. Modifier Additions Most aluminum foundry ingot is modified before shipment to the foundry. However, modification is subject to fade, as with grain refinement, and additional modification is therefore usually necessary. Hypoeutectic alloys are modified by additions of sodium, strontium, calcium, or antimony, as described below. Sodium is usually added in metallic form, encapsulated in an aluminum can to retard its natural burning in air. The sodium should be plunged into the melt and stirred gently for a short time to provide good dissolution and dispersion. Alternatively, several commercial salt flux tablets are available whose decomposition yields sodium for modification. The usual precautions apply regarding proper storage and the use of clean, dry tools. Fade, or loss of modification, increases with excessive agitation, degassing, prolonged holding periods, or excessively high temperatures (Ref 6). In general, a sodium addition of 0.015 to 0.020% is required for adequate modification, which results in a residual level of sodium in the casting of about 0.002%. Sodium should not be used with alloys containing more than 1% Mg, because of possible embrittlement. Strontium presents no special storage or handling problems, as does sodium, and is usually added in the form of Al- 10Sr-14Si or aluminum-strontium master alloy ingots or waffles. An Al-10Sr rod is also available, which dissolves more easily. Additions of 0.01 to 0.02% in foundry castings are sufficient, particularly when melting returns where typical residual 0.008% Sr is available. Overall, strontium modification seems more resistant to fade than sodium. Calcium is also added as a modifier in the form of a master alloy that also contains 5% Si. Nominal addition levels for modification are about 0.01%. Calcium also seems more resistant to fade than sodium, although prolonged holding will result in calcium loss (Ref 120). Antimony is a modifier that is used predominantly in Europe and Japan. It is added at about the 0.12% level and serves as a permanent alloy constituent of the aluminum alloy. It is therefore added by the supplier of foundry ingot and does not require any make up addition at the foundry. Antimony is reported to be unaffected by holding time, remelting, or degassing. Alloys modified with antimony are distinguished by lower gas susceptibilities than those containing sodium or strontium. However, cross contamination problems are possible, and antimony is more toxic. Hypereutectic Alloys. Although various elements have been found to be capable of modifying (refining) hypereutectic alloys, only phosphorus is used commercially. Other elements, such as magnesium, tungsten, sulfur, and lanthanum, have been reported to be effective modifiers (Ref 113) but thus far none has achieved commercial importance. Phosphorus is available for the refinement of hypereutectic alloys in the form of master alloys (phoscopper), aluminum phosphide, or silicon phosphide; as phosphorus pentachloride; or as various proprietary salt mixtures. Additions usually range from 0.020 to 0.025% P. Addition should be made carefully, with the customary precautions to ensure completely dry materials and tools. Modifier Comparisons and Interactions Dissolution. With the advent of strontium modification for hypoeutectic alloys and the increased use of foreign castings modified with antimony, many questions have arisen concerning the merits, problems, and interactions between sodium, strontium, and antimony. Strontium modification has been examined in several recent papers, and the nature of the strontium addition has been found to be quite important (Ref 121, 122, 123). An oxide-free strontium surface increases the rate of dissolution and minimizes the so-called hydrogen pickup that has sometimes been associated with strontium additions (Ref 122). In general, the strontium master alloys perform better than strontium-bearing salts, and sheared rod produces faster dissolution than waffle ingot. Strontium dissolution is characterized by an exothermic reaction that results in the formation of intermetallic compounds (Ref 121, 122). The rate of this reaction, which is temperature dependent, determines the effectiveness of strontium recovery and therefore the degree of modification achieved. It has been found that when elemental strontium is present, as in a 90Sr-10Al master alloy, the exothermic reaction proceeds at about 725 °C (1340 °F). However, at lower temperatures, the reaction is slowed, and an intermediate intermetallic precipitate, SrAl 2 Si 2 , forms with good dissolution characteristics. Therefore, lower temperature provides better recovery than higher temperature (Fig. 76a). Fig. 76 Comparison of dissolution rates for strontium additions made with 90Sr-10Al (a) and 10Sr- 90Al (b) master alloys. Source: Ref 123. On the other hand, lower strontium master alloys such as 10Sr-90Al dissolve by simple dissolution; therefore, recovery increases at higher temperatures (Fig. 76b) because the dissolution rate is greater (Ref 123). A certain amount of time (10 to 20 min) may be necessary for optimal recovery, which is dependent on temperature and the specific strontium alloy addition (Ref 124). Sodium dissolution occurs more simply and rapidly, providing almost immediate optimal modification. Antimony appears to behave similarly. All modifiers increase the eutectic suppression. At equivalent levels of suppression, strontium yields better modification than sodium. Fade. It is well known that sodium modification fades; that is, sodium is lost through oxidation and reaction within the melt, perhaps forming some sodium-aluminum-silicon complex that retards the ability for sodium to modify. Degassing or melt agitation easily destroys sodium modification, while such stirring does not seem to affect strontium modification (Ref 125). Contamination. The recycling of scrap castings poses the problem of intermingling of different modifiers and possible detrimental consequences. Sodium and strontium have been found to be compatible with each other in the range of 0.007 to 0.012% Na and 0.007 to 0.019% Sr (Ref 124, 126). The strontium presence overcomes the effect of sodium fade for up to nearly 2 h. Therefore, strontium can be used to modify melts that may contain some sodium from recycled scrap. Antimony, however, poisons both sodium and strontium modifications. In the case of sodium, at normal sodium addition levels near 0.01%, the presence of 0.02 to 0.05% Sb eliminates modification. A level of 0.01% Sb might be tolerated, but requires up to 0.04% Na to avoid fade and to retain modification after 1 h (Ref 126). It is thought that this negative interaction is a result of a sodium-antimony (possibly Na 3 Sb) intermetallic compound that forms, depleting the alloy of sodium available for modification. Antimony has a similar negative effect on strontium modification. Figure 77 shows a change in antimony content when it is present as a contaminant in strontium-modified melts. There is reasonably clear evidence that a Mg 2 Sb 2 Sr intermetallic forms concurrently as the strontium dissolves. This precipitate, having a higher density than molten aluminum, forms a sludge at the bottom of the melting vessel. Figure 78 indicates that much higher amounts of strontium are needed to achieve modification when antimony is present. Fig. 77 Interaction between antimony and strontium present in A356 alloy as a function of time. Source: Ref 126. Fig. 78 Effect of antimony on strontium modification in A356 alloy. Source: Ref 126. When all three modifiers are present, the ternary interactions behave as expected on the basis of the binary interactions. The antimony content decreases as either strontium or sodium is added, reflecting the intermetallic compound formation. Sodium content decreases rapidly at first, especially when elemental sodium is plunged. If the total modifier concentration is high enough, particularly strontium at 0.010 to 0.020%, the effects of antimony contamination can be overcome and modification maintained. Phosphorus is chemically similar to antimony and has a similar deleterious effect on either sodium or strontium modification. For a given level of phosphorus (which could be present as an impurity arising from hypereutectic alloy scrap or phosphate-bonded refractory materials), a higher amount of either sodium or strontium is necessary to achieve full modification. Hydrogen and Degassing. The effect of modification practice on hydrogen gas absorption and the subsequent response in degassing have been commonly observed and studied in some detail. Researchers (Ref 124) have noticed a tendency toward higher gas levels using either sodium or strontium (Fig. 79). The increase in gas content does not necessarily occur immediately, but increases with time. An increase in gas absorption when modifiers are present is especially noticeable at temperatures above 745 °C (1375 °F). Deliberately gassed strontium-modified melts show greater hydrogen uptake and less natural outgassing than unmodified melts. The influence of antimony on gas levels has not been discussed in the literature as frequently, but antimony modification appears less gas prone than strontium modification (Ref 127), with much higher densities (less porosity) reported for a given gas content (Fig. 80). [...]... Al2O3 Particles, skins 3.97 0.143 0. 2-3 0 (particles) 1 0-5 000 (skins) 8- 1 200 40 0-2 × 105 Magnesia MgO Particles, skins 3. 58 0.129 0. 1-5 (particles) 1 0-5 000 (skins) 4-2 00 40 0-2 × 105 Spinel MgAl2O4 Particles, skins 3.60 0.130 0. 1-5 (particles) 1 0-5 000 (skins) 4-2 00 40 0-2 × 105 Silica SiO2 Particles 2.66 0.096 0. 5-5 2 0-2 00 Chlorides Varies Particles 1.9 8- 2 .16 0.07 2-0 .0 78 0. 1-5 4-2 00 Fluorides Varies Particles... 0.07 2-0 .0 78 0. 1-5 4-2 00 Fluorides Varies Particles 1.9 8- 2 .16 0.07 2-0 .0 78 0. 1-5 4-2 00 Aluminum carbide Al4C3 Particles 2.36 0. 085 0. 5-2 5 2 0-1 000 Silicon carbide SiC Particles 3.22 0.116 0. 5-2 5 2 0-1 000 AlN Particles, skins 3.26 0.117 1 0-5 0 40 0-2 000 Titanium boride TiB2 Agglomerated particles 4.50 0.163 1-3 0 4 0-1 200 Aluminum boride AlB2 Particles 3.19 0. 115 0. 1-3 4-1 20 Salts Carbides Nitrides Aluminum nitride... and R.A Flinn, The Effects of Tin, Aluminum, Nickel, and Iron on Dissolved Gases in 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 Molten Copper Alloys, Trans AFS, 1 980 , p 437 Casting Copper Base Alloys, American Foundrymen's Society, 1 984 T.R Ostrom, R.A Flinn, and P.K Trojan, Gas Content of Copper Alloy Melts Test Equipment and Field Test Results, Trans AFS,... Eng., March-April, 19 78, p 22 R.W Bruner, Basics on the Fluxing of Aluminum, Die Cast Eng., Nov-Dec 1 986 , p 42 S.C Jain and S.C Tiwari, Fluxing Processes in Production of Aluminum and Its Alloys, Indian Found J., Vol 25, 1979, p 1 2-1 6 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 Aluminum Casting Technology, American Foundrymen's Society, 1 986 E.F Emley,... America, 1 986 R.C Harris, Deoxidation Practice for Copper, Shell-Molded Castings, Trans AFS, 19 58, p 69 J.L Dion, A Couture, and J.O Edwards, "Deoxidation of Copper for High Conductivity Castings," Report MRP/PMRL-7 8- 7 (J), Physical Metallurgy Research Laboratories, CANMET, April 19 78 B.L Tiwari, Demagging Processes for Aluminum Alloy Scrap, in Light Metals, The Metallurgical Society, 1 982 , p 88 9 G Dube,... Trans AFS, 1 985 , p 291 J.E Gruzleski, N Handiak, H Campbell, and B Closset, Hydrogen Measurement by Telegas in Strontium Treated A356 Melts, Trans AFS, 1 986 , p 147 N Handiak, J.E Gruzleski, and D Argo, Sodium, Strontium, and Antimony Interactions During the Modification 127 1 28 129 130 131 132 133 134 135 136 137 1 38 139 140 141 142 143 144 145 146 147 1 48 149 150 151 152 153 154 155 156 of (A356)... for example (Ref 145) Fig 87 Typical bonded particle filter Fig 88 Schematic of a vertical filter (bonded particle) used for holding furnace filtration for die casting operations One proprietary bonded particle filter system employs the bonded particle filter in a three-dimensional configuration (Fig 89 ) This system has been successfully used in mill product, foundry, and die casting operations The filter... Magnesium Alloys, Trans AFS, 19 68, p 92 H.I Kaplan, Basic Metallurgy of the Magnesium Die Casting Alloys, Die Cast Eng., Nov-Dec, 1 986 Magnesium Die Casting Manual, Dow Chemical Company, 1 980 J.F Wallace and R.J Kissling, Fluxing of Copper Alloy Castings, Foundry, 1963 J.F Wallace and R.J Kissling, Gases in Copper Base Alloys, Foundry, Dec 1962, p 3 6-3 9; Jan 1963, p 6 4-6 8 L.V Whiting and D.A Brown, "Air/Oxygen... for Al-Si Casting Alloys, Trans AFS, 1 987 M.O Pekguleryuz, B Closset, and J.E Gruzleski, The Dissolution of Metallic Strontium in Liquid Aluminum and Liquid A356 Alloy, Trans AFS, 1 984 , p 10 9-1 18 M.O Pekguleryuz and J.E Gruzleski, Conditions for Strontium Master Alloy Addition to A356 Melts, Paper 15, Trans AFS, 1 988 T.J Hurley and R.G Atkinson, Effects of Modification Practice on Aluminum A-356 Alloys,... Aluminum Castings, in Proceedings of the International Molten Aluminum Processing Conference (City of Industry, CA), American Foundrymen's Society, 1 986 , p 219 1 18 T.J Hurley, Using Electrical Conductivity and Ultrasonics to Determine Modification in Al-Si Alloys, Trans AFS, 1 986 , p 159 119 B Closset, Modification and Quality of Low Pressure Aluminum Castings, Paper 76, Trans AFS, 1 988 120 A March, High Volume . 1.9 8- 2 .16 0.07 2-0 .0 78 0. 1-5 4-2 00 Fluorides Varies Particles 1.9 8- 2 .16 0.07 2-0 .0 78 0. 1-5 4-2 00 Carbides Aluminum carbide Al 4 C 3 Particles 2.36 0. 085 0. 5-2 5 2 0-1 000 Silicon. Al 2 O 3 Particles, skins 3.97 0.143 0. 2-3 0 (particles) 1 0-5 000 (skins) 8- 1 200 40 0-2 × 10 5 Magnesia MgO Particles, skins 3. 58 0.129 0. 1-5 (particles) 1 0-5 000 (skins) 4-2 00 40 0-2 ×. MgAl 2 O 4 Particles, skins 3.60 0.130 0. 1-5 (particles) 1 0-5 000 (skins) 4-2 00 40 0-2 × 10 5 Silica SiO 2 Particles 2.66 0.096 0. 5-5 2 0-2 00 Salts Chlorides Varies Particles 1.9 8- 2 .16