Thomas, B.G., "Metals Processing", Chapter 14 in Structure, Processing, and Properties of Engineering Materials, J. Adams, editor, Addison Wesley, in press. Ch. 14 Metals Processing 1. Introduction 2. Metal Extraction 3. Casting A. Steps in Solidification B. Casting Processes C. Casting Defects D. Casting Process Analysis E. Summary of Casting 4. Thermal Processing A. Precipitation Hardening B. Welding C. Summary of Thermal Processing 5. Mechanical Processing A. Cold Working B. Hot Working C. Mechanical Deformation Processes D. Summary of Mechanical Processing 6. Summary 2 Ch. 14 METALS PROCESSING 1. INTRODUCTION Even if an engineer is not directly involved with materials processing, it is important to know something about how materials are made into parts. A material is not useful until it is processed into a desired shape. Furthermore, the final shape, structure and properties of a component are determined by its processing. Processing can be costly, so the possible shapes and properties of a low-priced part are limited. In addition, processing may be the source of defects in the final product. One of the advantages of metals is the wide range of economical processes available to alter their shape and properties. The basic processes of extraction, casting, thermal treatment, and deformation are discussed in turn in this chapter in the context of metals. In addition to these processes, manufacturing also involves joining parts together by a variety of methods including welding, brazing, soldering, fasteners, and adhesives. Their shape and surface finish can be altered by mechanical and chemical treatments, such as machining and plating. These are important operations but are beyond the scope of this text. Steel Production: a case study in metal processing The manufacture of steel requires a long sequence of processing steps, shown in Figure 14.1. Steel production starts with making iron in a blast furnace. Iron ore (most commonly Fe O ) is 2 3 charged (added) with coke (carbon) and limestone (mainly CaCO ) to the top of the tall, shaft- 3 like, furnace. Air is blown through the furnace to burn the coke to provide heat as well as CO gas, which reacts with the oxygen in the iron ore to release the iron atoms as liquid metal. The limestone reacts with the waste rock in the ore to form a low-melting slag that floats on top of the iron. The molten (liquid) iron is tapped (removed) from the bottom of the furnace. This iron is saturated with carbon, (over 4% C) and also absorbs other impurities such as S, P, and Si from the ore, limestone, and coke. This impure product must be refined to make steel. To do this, the metal is usually poured into a large (250 ton) vessel called a Basic Oxygen Furnace, along with scrap iron and slag-forming materials. Bubbling oxygen through the iron bath for a short time (about 30 min.) takes away most of the unwanted carbon as CO bubbles, transforming the iron into steel (0 - 1%C). Other refining stages and alloy additions are made to remove the impurites and adjust the final composition. For example, Al or Si is added to combine with the unwanted 3 excess oxygen. The finished steel is poured into a ladle and then cast. In the past, permanent molds were used to cast large, steel shapes called ingots. The more efficient process of continuous casting has recently transformed the steel industry. In this process, the steel moves through a holding vessel, called a tundish, and down through a continuous casting machine, where it solidifies into long rectangular slabs, or square billets. These semi-finished shapes are processed by sequences of thermal cycling (eg. annealing and cooling), chemical baths, and deformation processes (eg. hot and cold rolling) to produce finished shapes with a wide range of shapes, structures and properties. This procedure is used to produce 90% of the world’s steel, efficiently enough for it to be sold for pennies per pound. Similar processes are used to make most other commercial metals. Further details on the many complex thermal processes, or heat treatments, possible in steel are described in the next chapter. Key Concepts (cid:127) extraction - production of liquid metal (cid:127) casting - solidification of liquid metal into a solid shape (cid:127) thermal treatment - sequences of heating and cooling to alter structure and properties (cid:127) deformation processing - changing shape, structure, and properties via mechanical working 2. METAL EXTRACTION The first step in producing metallic parts is to extract the metal from its ores, which usually creates liquid metal with the desired composition. This stage is important because, if not done carefully, the liquid metal can contain too many impurities, detrimental inclusion particles or dissolved gases, which can initiate defects later. An ore is a mineral deposit from which metal can be economically extracted by chemical reduction. There are two fundamental approaches to the extraction of metals: 1) produce a relatively impure metal from the ore, which is then refined, or 2) produce a purified mineral from the ore, which is then reduced into a pure metal. Steelmaking is an example of the first approach, described in Figure 14.1. The production of aluminum from its ore, bauxite, (mainly Al O ) is an example of the second 2 3 approach. By a series of solution and precipitation steps, a very pure aluminum oxide is produced, which is then melted and reduced to metallic aluminum by electrolysis. Enormous amounts of electricity are required for this process, which makes the recycling of aluminum environmentally as well as economically sound. 4 After the metal is produced, further processing steps are usually required before it is ready for manufacturing. These steps include refining, which removes impurities from the molten metal, and alloying, which adjusts its composition. ______________________________________________________________________________ Example 14.1 - Refining Gases dissolved during the extraction of molten metal can be removed by vacuum degassing, in order to avoid defects during later processing, such as porosity. (See Section 14.3C). How much dissolved hydrogen gas could be removed from molten aluminum at 700oC by passing it through a 10-4 atm vacuum chamber? Assume the initial ambient environment contains 0.1 atm H (due to high moisture content of the air). 2 Answer: First, find the amount of gas that can be dissolved in the molten metal, which is given by Sieverts’ law, which is plotted in Figure 14.2 for hydrogen in aluminum: dissolved gas = K p gas where p is the partial pressure of the gas in the environment above the liquid (atmospheres) gas and K is an empirical equilibrium constant (wt. ppm at 1 atm) that increases with temperature: K(H in Al)=505exp −6255  T(ÞK)  -6255  So, at T = 700oC, K= 505 exp   = 0.816 700+273 and the initial equilibrium H content, [H] = K p = 0.816 0.1 = 0.26 ppm H in Al. gas Next, find the maximum H content in the refined Al produced by the partial vacuum: [H] = 0.816 0.0001 = 0.008 ppm H removed by this high vacuum = 0.26 - 0.008 = 0.25 ppm. Thus, most of the original dissolved gas can be removed, if sufficient time is allowed to reach equilibrium. Care must be taken to avoid exposing the refined metal to air or moisture after the degassing treatment, or gas may redissolve. Porosity defects will form during solidification if aluminum contains more than 0.1 wt. ppm dissolved H (1 ml H gas / kg). 2 5 ______________________________________________________________________________ 3. CASTING The next manufacturing step is to create the basic shape. For almost all metals, this is done by casting: liquid metal is poured into a mold cavity, where it transforms, or solidifies into a solid of the desired shape. Sometimes casting produces the final part. Sand casting, for example, is one of the oldest manufacturing processes, dating back thousands of years, and is often still the most economical one. This process is the only way to make certain large complex metal shapes, such as 100 ton turbines for hydro-electric dams, or to manufacture parts from brittle materials, such as gray cast iron. Usually, the initially-solidified metal shape is an intermediate product requiring further processing steps before becoming a finished part. These steps include remelting and recasting (to change its shape), thermal treatment (to change its properties) and / or deformation (to change both). Because the manufacturing of metal parts almost always includes a casting process, it is important to understand the fundamentals of solidification and the inherent structure and defects that it can create. 3A. STEPS IN SOLIDIFICATION Solidification occurs in several stages: 1) nucleation , or formation of tiny solid crystals, 2) growth of these crystals into larger crystals called dendrites, and 3) further growth into grains, which make up the final solidified structure. Nucleation and Growth The first step of metal solidification is the creation of tiny, stable, solid crystals, or nuclei in the liquid metal. Cooling the liquid below its equilibrium freezing temperature, or undercooling, provides the driving force for solidification. For the new solid to be stable, however, the energy released in forming the new solid volume must be larger than the energy needed to create the new solid / liquid interface. Larger clusters of solid atoms are more stable than small ones, since they have less surface area per unit volume. Once a cluster reaches a critical size, it becomes a stable nucleus and continues to grow. The mold walls and any solid particles present in the liquid make 6 nucleation easier. They do this by providing solid surfaces for liquid atoms to solidify upon, which reduce the undercooling needed to less than 1oC. Each nucleus eventually grows into a single grain, whose shape depends on how neighbouring grains impinge upon each other. Thus, greater numbers of nuclei produce a finer (smaller) grain size. A finer grain size is generally beneficial to strength and toughness in the final product. Nucleation is enhanced by adding alloying elements or other materials (called grain refiners) that form small solid particles in the liquid. The seeding of clouds to precipitate rain is a similar process. Dendrites A casting usually begins with rapid nucleation and growth against the cold mold walls. This produces a thin chill zone of tiny grains at the casting surface. This shell of solid metal grows into the remaining liquid as solidification continues. Due to the combined effects of alloy segregation during freezing, slow diffusion and shallow temperature gradients in the liquid, the solid / liquid interface usually does not remain planar. Instead, tree-shaped spikes called dendrites shoot into the liquid in the direction of heat flow. The dendrites also grow short, perpendicular secondary arms, as seen in Figure 14.3. Dendrites create a mushy zone between the solid shell and the liquid, bounded roughly by the liquidus and solidus temperature contours. Certain crystallographic directions in the solid metal tend to grow faster than others. Thus, some dendrites grow faster than others into the direction of heat flow. The result of this competition is an array of parallel aligned dendrites growing away from the randomly-oriented chill zone. The parallel dendrites eventually form a large region of grains with a similar orientation, perpendicular to the mold wall, called the internal columnar zone. These columnar grains are very long and thin, with long parallel grain boundaries that can be a source of weakness in the final casting. Grain Structure While columnar grains are growing inward from the mold walls, other nuclei are simultaneously growing in the central liquid pool. These central nuclei survive because the first solid to freeze in a typical commercial alloy has a lower alloy content, and hence a higher melting point, than the surrounding liquid. The central crystals then grow equally in all directions, forming roughly- round-shaped, equiaxed grains. Eventually, the growth of the columnar grains is stopped when 7 they impinge on the equiaxed grains. The result is a macrostructure consisting of three zones, as shown in Figure 14.4. The relative proportions of the chill, columnar, and equiaxed zones depends on the alloy and thermal conditions during solidification. Preheated molds decrease the depth of the chill zone. Conditions that produce large numbers of nuclei in the central zone tend to result in a large equiaxed zone with a smaller average grain size. A lower pouring temperature promotes a larger equiaxed zone, by making it easier for nuclei to survive in the center. Alloys with a wide freezing range, or large difference between the solidus and liquidus temperatures, similarly promote equiaxed grains. This is because dendrite arms may be broken off by stirring and transported to the center, where they are able to avoid remelting, due to their different composition, and act as nuclei. Castings that consist mainly of uniform equiaxed grains usually have the best mechanical properties. ______________________________________________________________________________ Example 14.2 Solidification Structure How does reducing the secondary dendrite arm spacing (SDAS) of an aluminum alloy from 0.1 mm to 0.05 mm affect the strength of the casting? Answer: The tensile strength, UTS, generally increases with decreasing SDAS. The specific relation for aluminum is: UTS (MPa) = 48.5 - 90 SDAS (mm) So, Strength of large SDAS = 48.5 - 90 (.1) = 40 MPa Strength of small SDAS = 48.5 - 90 (.05) = 44 MPa Thus, the strength increases from 40 to 44 MPa with this decrease in SDAS. Mechanical properties are improved by casting processes which refine the microstructure, by decreasing the SDAS and the grain size. ______________________________________________________________________________ 3B. CASTING PROCESSES Continuous casting involves pouring liquid metal through a bottomless permanent metal mold. The thin shell that solidifies against the mold walls is pulled continuously out of the bottom and acts as a container for the remaining liquid. (See Figure 14.1) As the soft shell is withdrawn, it 8 is spray cooled with water and may need to be supported between rollers. The shell continues to solidify and thicken until the cross section of the casting is fully solid. The final strand of metal leaving the machine is cut with a torch to the desired lengths. This process is the most efficient way to solidify large volumes of metal into simple shapes for subsequent processing. It can be applied to most metals, including steel, with a variety of sizes and shapes. Cross-sections can be rectangular, square, circular, and even “dog-bone” shapes, for rolling into I-beams. Most of the 500 million tons of steel produced in the world each year is continuously cast. Sand casting uses a two-piece mold formed by packing sand around a wood or metal pattern that is shaped like the intended part. (See Figure 14.5) The sand is held together by a binder, consisting of oil or clay. A gating system is designed to let liquid metal flow into the mold cavity as fast as possible with a minimum of turbulence. The pattern also includes features to minimize defects, such as risers and chills that control heat transfer (see Section 14.3C). A recent development is the molded styrofoam pattern, which does not have to be removed before casting. Sand casting is the most common, versatile, and economical casting process for manufacturing parts, especially in small numbers. It is also one of the few processes that can mass produce large ferrous parts, such as automotive engine blocks. Since the sand mold must be remade for each casting, the process is labor intensive. Permanent mold casting, also called gravity die casting, is similar to sand casting with the advantage of a reusable metal mold. Before continuous casting, permanent cast-iron molds were popular for making large semi-finished steel shapes, called ingots. Now, the process is used mainly for small, low melting point castings (eg. Al or Mg), when hundreds of identical castings are needed. Casting hotter metals such as steel into finished parts is usually not economical because the metal attacks the mold and shortens its life. Die casting involves injecting liquid metal under pressure into a permanent metal mold. The two pieces making up the mold, called dies, are clamped together prior to metal injection and open quickly after solidification to achieve high production rates (100 parts per hour per cavity). The process is similar to injection molding used in ceramic and plastic processing, shown in Figures 16.11 and 17.8. 9 Die casting can rapidly produce shaped parts with high dimensional accuracy. The dies are very expensive, but are cost-effective for mass production, since they can be reused thousands of times. Like gravity die casting, this process is limited to relatively small shapes and to nonferrous alloys with low melting temperatures. Investment casting also called the lost wax method, starts with a wax pattern of the desired , shape. The pattern is dipped repeatedly in a ceramic slurry to form a thin coating. After melting out the wax, the ceramic is usually fired (baked). The strong ceramic shell that remains is the mold for casting the liquid metal. Investment casting reproduces fine detail very well and achieves excellent surface finish and dimensional accuracy. Its high cost limits the process to small quantities of small castings such as jewelry, dental work, and gas turbine blades. Directional solidification is a modification of the investment casting process, designed to control solidification to progress only in a single casting direction. One way to accomplish this is to slowly withdraw the mold, filled with molten metal, down through a hole in the floor of a two- zone vacuum furnace, shown in Figure 14.6. Solidification occurs only at the junction between the hot (upper) zone and the cold (lower) zone. This process aims to avoid the formation of grain boundaries, particularly those perpendicular to the casting direction. If solidification progresses from a single nucleus, then a single crystal casting is formed. This is often promoted by placing a grain selector at the bottom of the mold, which forces solidification through a spiral-shaped passageway. Only the grain with the most favorable orientation survives. The advantage of producing single crystal parts (with no grain boundaries) is improved creep strength at high temperature, as discussed in Chapter 10. This is important for parts such as rotating, high- temperature, jet-engine turbine blades, which are often cast as single crystals, despite their complex shape (See Figure 14.7). Other types of directional solidification processes are used to purify metals (zone refining) or to make pure, single-crystal bars such as silicon semiconductor material. Other casting processes exist for making specialized parts. For example, hollow pipes can be made by centrifugal casting, where metal is poured into a rotating cylindrical mold. Vacuum casting processes, such as ESR (electro-slag remelting), are used to remelt and resolidify metal in order to remove inclusions and reduce segregation for parts with critical applications. Other parts can be made by squeezing them into shape from semi-solid metal. Metal droplets can be spray cast into powders and later sintered together into shaped parts in a similar manner to the way ceramics are produced. Thin metal sheets can be produced by casting onto a rotating wheel, between rotating wheels, or onto moving belts. The high cooling rates possible in some of these 10 processes achieve highly non-equilibrium conditions, which increase the range of microstructures and properties possible. Development of new casting processes such as these is the subject of intense current research. ______________________________________________________________________________ Example 14.3 Casting Processes What casting process would you use to manufacture steel I-beams for building construction? Answer: Continuous casting of “beam blanks” is the most efficient way to produce long shapes, which have a uniform cross-section, shaped like a dog-bone in this case. These semi-finished shapes are later rolled to their exact shape and dimensions to produce I-beams. Example 14.4 Casting Processes What casting process would you use to mass produce steel camshafts for auto engines? Answer: Foundry (sand) casting is the most practical approach for steel parts. Die casting would be more efficient, but is not economical for steel, because the melting point is so high. ______________________________________________________________________________ 3C. CASTING DEFECTS Casting may produce internal and surface defects. If strength and toughness are critical and cost is not of overriding importance, then castings are often subjected to further processing to improve their properties. Many of the problems that occur in finished products have their origin in the original casting. Internal defects created during casting can be very costly, since they are usually difficult to detect, and are often not found until later processing stages or even in service. Some defects are caused by simple mistakes, but these are specific to the individual casting process. The following generic problems are always a concern. Inclusions Nonmetallic particles, or inclusions, can be created by oxidation of the liquid metal, or particles entrained from slag, dirt, or refractories when the metal is fed into the mold. These particles, such as alumina (Al O ), cannot deform as metal does, owing to their brittle ceramic structure. 2 3 Large inclusions are serious defects. For example, the production of beverage cans is limited by the number of inclusions in the metal, since every particle bigger than about half the wall thickness will cause a failure. Large inclusions can also initiate cracks when a part is in service. (See Chapter 9). Inclusions can be minimized by preventing exposure of the liquid metal to air,