The United States is the world's largest consumer and producer of aluminum. In 1995, the primary smelting sector of the industry produced 3.375 million metric tons of aluminum, constituting 17.3 percent of world production. In addition, the secondary refining sector recovered 3.188 million metric tons of recycled aluminum. Imports made up the rest of the 9.265 million metric tons of aluminum consumed in the United States that year (Aluminum Association, 1997). Total annual shipments of aluminum in 1993 were valued at $21 billion, and the industry employed 134,000 workers (OIT, 1997c).
Aluminum is an energy-intensive industry, with an annual energy use of 1 quadrillion Btus and energy expenditures of $2 billion, or 8.6 percent of the total value of shipments in 1993 (OIT, 1997c). Production of aluminum from recycled materials requires only 5 percent as much energy as primary production and thereby decreases the total energy consumption of the industry.
Various sectors of the aluminum industry, from primary smelters to producers of finished products, have diverse separation needs. In terms of basic processes, the industry needs improvements in separation processes that increase metal purity. The industry as a whole has a need for innovative separation technologies to improve the sorting, and thereby the quality, of scrap. The recycling of scrap reduces energy use, reduces costs for the purchase of raw materials, and reduces the impact of waste on the environment. The separation needs of the aluminum industry are shown in Figure 5-1.
Separation Processes for Improving Metal Purity
Separation processes are an important part of the aluminum manufacturing process. The three basic processes used in making aluminum are the Bayer process, for the conversion of bauxite into alumina, the Hall-Heroult process, for the electrolytic smelting of alumina to aluminum, and the alloying, casting, forging, and fabrication processes for making products. These basic processes are relatively straightforward. The industry has an overriding, constant need to maintain the purity of material flows because most other metals are more noble than aluminum. Once an impurity is introduced into the aluminum melt, it is practically impossible to remove it. Therefore, the industry's standard practice is to avoid impurities as much as possible.
The Bayer process is the principal method for producing alumina feedstock. In this process, bauxite ore, a mixture of aluminum, iron, titanium, and silicon oxides, is crushed and milled and then mixed with a concentrated solution of sodium hydroxide. This mixture is fed into autoclaves where the alumina in the ore is dissolved to form sodium aluminate at temperatures of up to 270°C. Silica reacts to form sodium aluminum silicate, which precipitates out of solution, and the other oxides remain as solids. The sodium aluminate is recovered separately and converted to alumina. The solid-rich residue, known as red mud, is washed to recover sodium hydroxide and then disposed of.
Because of the retention of sodium oxide in the Bayer process, aluminum fluoride must be added to the electrolytic cell, which results in the generation of excessive cryolite. Minor impurity levels of other oxides, such as iron, zinc, and silica, are reduced in the electrolytic cell and become major impurities in the metal. Calcium oxide, if retained, is a major impurity in the electrolyte. The aluminum industry would benefit from improved separation technologies to remove these impurities from the metal and the electrolyte, or to avoid them altogether.
During the Hall-Heroult process, the electrolytic cells generate solid and gaseous fluorides, as well as CO2 emissions. All smelters use scrubbing systems to treat the off-gases. Dry scrubbers, which are designed to put the gases in contact with alumina in dispersed phase or fluidized bed reactors, are the preferred technology because unreacted alumina is used to capture hydrogen fluoride and electrolyte compounds that can be returned to the cell. Elemental impurities (e.g., fluorine, vanadium, lithium, iron, phosphorus, and nickel) concentrated in the scrubber alumina are subsequently returned to the cell. Once in the cell, iron reduces metal quality, phosphorus reduces current efficiency, and vanadium reduces both. Therefore, better separation technologies for removing iron, phosphorous, and vanadium from the scrubber alumina would be of considerable interest to the industry.
During cell tapping, electrolyte is inevitably removed along with the metal and ends up in casting centers as a dross contaminant. It can have an additional detrimental effect through back reaction with aluminum to produce elevated levels of light metals in the melt, which in turn creates a need for additional chlorine fluxing.
Chlorine gas fluxing is used in combination with argon or nitrogen to remove light metal impurities, such as sodium and lithium, hydrogen gas, and oxide inclusions, from molten aluminum. Chlorine fluxing is also used extensively in the secondary, or recycling-based, sector of the industry to selectively remove unwanted magnesium from casting alloy melts. The magnesium is present in the melt because a mixture of scrap, including both cast and wrought alloys, is used for the melt, and the wrought alloys contain magnesium levels that exceed the cast alloy target specification. The industry would benefit from alternatives to removing impurities via chlorine fluxing, which is expensive and creates environmental problems.
Filtration of molten metal is a complementary technique to fluxing for removing oxides and nonmetallic inclusions from the melt. Filtering is designed to provide consistent metal quality for a wide range of casting requirements. The techniques employed are ceramic foam filters (which use replaceable filter cartridges), packed beds, and sedimentation. A better understanding of filtration mechanisms would increase the efficiency and reliability of this technology, as well as lower the costs. The industry also needs new techniques for sensing inclusions in metal.
In the Hall-Heroult process, a direct current is passed through a current-conducting bath of molten cryolite (Na3AlF6) in which alumina is dissolved. The bath is contained within an insulated cell. Carbon anodes are suspended above the cell and are dipped into the bath. The cathode is the molten aluminum. Under the influence of the applied current, aluminum is deposited at the negative electrode and collects at the bottom of the cell. Oxygen is released at the carbon anodes where it reacts to form CO and CO2 at the expense of the anode material. Approximately two tons of alumina and one-half ton of carbon are required for the production of one ton of aluminum. Hence the primary smelters are significant emitters of CO2. The carbon anodes are produced in a separate plant from a mixture of calcined petroleum coke and coal tar pitch, which has been prebaked.
Avoiding CO2 emissions and the search for nonconsumable electrolytic anodes are linked. CO2 is created by the current practice of using consumable carbon anodes in reduction cells, and the aluminum industry is searching for alternatives. The ideal electrode would be a nonconsumable inert material with the required current conducting properties and the ability to withstand the aggressive environment in the cell. Improvements in this Hall-Heroult method of reacting and separating out aluminum would reduce CO2 emissions.
After the electrolytic cell is taken out of service, spent potlining (the insulation and containment material of the electrolytic cell) remains. This material often contains fluoride compounds and sodium cyanide, which must currently be processed as hazardous waste. The industry would welcome better ways of dealing with
spent potlining, such as methods of separating out components before processing to reduce process costs and volume. Therefore, a method of removing hazardous materials from spent potlining would benefit the industry.
Identifying and sorting scrap was cited as a major need by every sector of the aluminum industry because of the diversity of aluminum alloy specifications and because of the restrictions imposed by those specifications on the chemistry of scrap remelts. The quintessential example is the recycling of secondary foundry alloys compared to the recycling of used beverage cans. Secondary foundry alloys are generally quite liberal in their tolerance for alloying elements, especially elements such as iron and silicon, which are routinely picked up during the recycling process. Silicon is a major component of foundry alloys because it improves melt fluidity needed for casting. Therefore, many secondary foundry alloys are able to accept wrought alloy scrap as melt ingredients. Magnesium levels in wrought alloy specifications often exceed foundry alloy specifications, but magnesium can be readily removed from the melt by chlorinating or by specific fluxes. In contrast, wrought alloys (with the exception of alloys such as 3105 utility sheet designed specifically to be made from recycled metal) are usually intolerant of impurities, and, except for the 4XXX series, are necessarily low in silicon. Therefore, although mixed wrought alloy scrap may be used to make secondary foundry alloys (with the use of fluxing where necessary to reduce magnesium), foundry alloy scrap cannot be used to make wrought alloys. The degree of tolerance among wrought alloys is entirely dependent on the chemistry of the target alloy. Used beverage cans, which are a mixture of 3104 (an aluminum alloy containing manganese) and 5182 (an aluminum alloy containing magnesium), are compatible for remelting because the combined melt chemistry is compatible with fresh 3104, allowing for the removal of excess magnesium through oxidation.
The global scrap aluminum balance is maintained today because of the healthy growth of the secondary foundry alloy business, particularly for automotive casting applications. Indeed, this industry consumes not only all of the old scrap castings and automobile scrap, which are a good mix of cast and wrought alloys, but also mixed wrought and cast alloy white goods, such as refrigerators and other household appliances. In the absence of this healthy demand, the mixed alloy scrap would have to be segregated to find a recycling outlet. The aluminum industry has targeted the automotive sector for significant growth, especially in the area of wrought alloy applications for structures and closures. If this growth is realized, aluminum intensive vehicles will contain as much, or more, wrought alloy as foundry alloy. The recycling of these vehicles will create a demand for separating component alloys into wrought and cast alloys, and subsequently into individual alloy streams,
to satisfy recycling and material supply criteria. This will require improved methods for separating aluminum scrap.
Dross is the oxide-rich by-product of aluminum melting operations. White dross from smelting or clean remelt operations may be rich in metal, with the balance aluminum oxide, whereas so-called black dross or salt cake contains salt flux, which is used to facilitate the wetting of aluminum scrap, as well as oxide and reduced amounts of metal. Salt cake, which is reactive, contains aluminum nitride, aluminum carbide, and fine aluminum and is a candidate for recycling to recover the entrained metal and salt flux. These recycling methods have yet to be developed.
Gaseous and Aqueous Emissions
The aluminum industry, in particular the remelt and recycling sector, is regulated for the emission of typical gaseous effluents, such as HCl, NOx, SOx, VOCs, and fine particulates in gaseous streams. Emissions are controlled via incinerators, acid gas neutralizers injected into the exhaust stream, and bag house filters with neutralizing precoats. As regulations become tighter, the industry is looking for more efficient methods for separating or treating these emissions. Improved dewatering methods (i.e., more efficient removal of water from process and effluent streams), especially in the Bayer process, is another area for research.
The panel identified a number of separation technologies with potential to meet the separation needs identified by the aluminum industry. For example, ceramic foam filters show promise for removing impurities from molten metal, and the industry would benefit from a better understanding of this technology. In addition, the search for alternatives to carbon anodes should focus on nonconsumable inert materials with the required current conducting properties that are capable of withstanding the aggressive environment in the cell (note that this may lead to a greater consumption of electricity).
With an ideal separation technology, closed-loop recycling of scrap metal could be achieved for complex products like automobiles. The metallurgical value of
scrap would be maximized by eliminating the wasteful practice of using chlorine to remove valuable alloying elements like magnesium. Some of the separation could be done by hand, using component database tools and hand-held analytical devices. However, for complex product forms like automobiles, much of the old scrap will probably be processed by shredding, the objective of which is to reduce the size of the scrap pieces and liberate the desired material from impurities. Unfortunately, the size reduction results in material that requires very high-speed sorting for the process to be economical. High-speed noncontact techniques have been identified that would be capable of differentiating all alloys of interest.
The successful separation of different types of aluminum alloys will require continued work on the fundamental mechanics of high-speed conveying, the positioning scrap pieces in sequential arrays before they enter analysis, and methods of physically sorting analyzed pieces by alloy type. Separation technologies for upgrading scrap include sorting methods based on physical properties, such as magnetic and electrical properties, densities, and melting points (see Chapter 10).
Recycling Salt Cake
In Europe, disposing of salt cake is prohibitively expensive because of the high costs of landfills, so it is substantially recycled. In some countries, the presence of salt cake in landfills has been banned by legislation. In established recycling practices, the salt cake is crushed and milled to recover recyclable aluminum, which is returned for melting, and the residue is dispersed in water to dissolve the salt. The solids are removed by filtration, and the salt is recovered from the brine by flash evaporation. The recovered salt is returned to aluminum smelters for reuse as flux, but the evaporation process is energy-intensive and costly. The solid nonmetallic by-products, which contain oxides together with aluminum metal and aluminum nitride, will form hydrogen and ammonia in the presence of moisture and heat if unreacted. Applications for reacted or stabilized nonmetallic byproducts have been found in the cement industry, but there may be other opportunities for further separation and upgrading of these by-products. Other by-products from aluminum processing, such as bauxite residue, red mud, and bag-house dusts, could also be separated into useful streams.
Gaseous and Aqueous Emissions
Inorganic membrane modules composed of cordierite honeycomb monolith coated with a microfiltration membrane (e.g., Al2O3) and vanadium-based catalysts could be used for removing SO2 from flue gas. Oxy-fuel burners are used in some sectors of the aluminum industry, chiefly in recycling furnaces. The principal
attraction of oxy-fuel is its intrinsic low NOx attributes compared with conventional fuel-air burners. However, the high cost of oxygen remains a significant barrier to the use of this technology. Lower-cost, hot oxygen would enable the implementation of additional oxy-fuel systems. In addition, the use of oxygen, as opposed to air, for combustion results in energy savings because the energy used to heat the nitrogen component of air, the thermal value of which is lost in the flue gas, would be saved.
The aluminum industry has a broad array of separation needs that affect all production processes, from primary smelting to the manufacture of finished products. Specific areas for improvement include:
- the separation of impurities from alumina feedstock
- the separation of impurities from dry scrubber alumina
- alternatives to chlorine fluxing for the removal of light metals and magnesium molten metal filtration
- methods for sensing inclusions
- decreasing or eliminating CO2 emissions by the development of nonconsumable electrodes
- methods of reducing and recycling spent potliner
- methods for segregating/sorting scrap by alloy
- separation/uses for nonmetallic by-products
- salt/water separation for efficient recycling of salt cake
- treatment of dilute gas streams
- lower cost oxygen for oxy-fuel burners
- efficient removal of excess water in the Bayer process