Still the Iron Age
Page 17
Material balances should also include hot air blast delivered through tuyères placed around the hearth’s perimeter (Fig. 6.4). Depending on the operating pressure and on the degree of oxygen enrichment, the mass of hot blast air is mostly between 1.25 and 1.5 t/t of hot metal, and oxygen enrichment ranges between 30 and 300 kg/t of pig iron. Besides the hot metal (pig iron), the output of the smelting process consists of slag and gases. BF is tapped by drilling 2–4 tapholes (the clay-filled openings at the base of furnace hearth, resealed by clay after casting) 8–14 times a day to remove molten iron and slag.
Figure 6.4 Tuyères of Baosteel’s (Shanghai) No. 4 BF on its completion day in 2005. Corbis.
Because of their very different densities (slag at 2.5 t/m3, pig iron at 7.2 t/m3), these two materials do not mix, and the slag, forming a supernatant on hot iron, is easily separated and channeled into a slag pit while iron is transported, in special torpedo cars or in massive ladles, to casting machines. Specific slag quantities vary depending on the amount of charged slagging materials, quality of ores, and efficiency of the smelting process, ranging mostly between 250 and 300 kg/t of hot metal, just a third or no more than half of the typical yield during the early 1950s (Lüngen, 2013; Schmöle, Lüngen, & Noldin, 2014). I will deal with slag’s postmill fate in the next chapter summarizing the environmental impacts of iron and steel industry.
Gases generated by the reduction of iron ores and production of hot pig iron are typically composed of up to 23% each of CO and CO2, and a few percent of H2, with nearly all of the rest being N2. Their density is thus about 1.4 kg/m3, and with volumes ranging mostly between 1150 and 1550 m3/t of hot metal, the material flow amounts to at least 1.6–2.2 t/t of pig iron. Their energy density is only about a tenth of the energy density of natural gas (mostly between 3.2 and 3.6 MJ/m3), but that is high enough to justify their capture and reuse. Reusing the top gas requires first the removal of the entrained dust, thus eliminating one of the most polluting by-products of formerly uncontrolled ironmaking.
Material flows in the primary iron smelting in BFs are obviously restricted by many stoichiometrical and physical requirements, but the fundamental identity of the process allows for some significant departures in terms of charge inputs. Coke always dominates the supply of carbon, but it can be partially substituted by powdered coal or by injections of liquid or gaseous hydrocarbons; iron can be charged as a raw ore, as pellets, or as sinter, or as a combination of any of two or three of these materials; and BFs can operate without any supplementary oxygen or with substantial enrichment. Differences in these material balances are easily appreciated by tabulating a few examples of specific material inputs.
The first comparison (Table 6.1) illustrates the range of BF charges attributable to differences in the form and quality of charged ores (lump, sinter, pellets, mixtures of two or all of these), furnace sizes, and productivities, their use of coal injection, and blast’s enrichment by oxygen. The table lists actual charges for specific BFs or typical rates used in model calculations, all normalized per tonne of hot metal: they include an example of a typical Belgian–Dutch–German operation (Danloy et al., 2008), a Swedish BF in Luleå (Ryman et al., 2004), a large Russian furnace, Severstal No. 5, and a large Ukrainian furnace, Mittal Steel No. 9 (Tovarovskiy, 2013), Czech (Besta et al., 2012) and Polish (Burchart-Korol, 2013) operations, and typical performances for the EU (IEA, 2010; Remus et al., 2013) and Japan (Morita & Emi, 2003; Nogami et al., 2006).
Table 6.1
Material Flows in Large BFs
Blast furnaces Iron ore Sinter Pellets Coke Coal Oxygen Slag
Belgiuma 1584
209 180 300 260
Belgiumb 1581
286 197 51 263
Netherlandsc
1580
300 200 320 240
Swedend
1367 327
56 166
Czech Rep.e 104 1283 247 504
396
Polandf
1307 250 428
303
EUg
1340 155 320 200 50 250
EUh 180 1088 358 359
150–347
USi 1628
480
255
USj 1524
486
220
Ukrainek
1345 295 474
102 367
Russial
50 1023 489 412 87 300
Japanm 280 1160 190 380 120 40 300
Japann
1317 356 446 57 25 287
Japan TGRo
1320 357 357
270 251
All values are in kg/t of steel, except oxygen, which is in m3/t.
aDanloy et al. (2008).
bLüngen (2013) (ArcelorMittal Gent A).
cGeerdes, Toxopeus, and van der Vliet (2009).
dWang et al. (2008).
eBesta et al. (2012); all coke.
fBurchart-Korol (2013).
gIEA (2010).
hRemus et al. (2013), weighted EU average.
iBurgo (1999).
jMorris, Geiger, and Fine (2011); iron ore includes 47 kg of BF slag with 20% FeO.
kTovarovskiy (2013).
lTovarovskiy (2013).
mMorita and Emi (2003).
nMorita and Emi (2003).
oNogami et al. (2006); TGR=top gas recycling.
(All values are in kg/t of hot metal.)
The second comparison lists material inputs and outputs of three large BOFs in the United States, Japan, and Spain (Table 6.2). Material flows for all natural gas-based DRI processes are similarly simple: 1.7 t of pellets and 0.18–0.24 t of CH4 (and 100–135 kWh of electricity) per tonne of hot iron, while coal-based processes charge 0.42 t of coke and 6 kg of lime (IEA, 2010).
Table 6.2
Material Flows in Large BOFs
Basic oxygen furnaces Hot metal Scrap Ore Fluxes Oxygen Slag Gases
US (Miller et al., 1998) 877.6 201.6 16.8 56.5 75.3 100.1 100.1
EU (Remus et al., 2013) 860 220 9.7 48.5 49.5–70 125.0 91.0
Japan (Morita and Emi, 2003) 1033.0 28.0 11.0 31.0 66.6 50.0 124.5
(All values are in kg/t of crude steel.)
Naturally, charged materials claim most of the attention in analyzing the balances of BF ironmaking and BOF steelmaking, but water is also an indispensable input. Three uses claim most of the supply: material conditioning (dust control in sintering, slag quenching in BFs, scale removal in hot rolling) consumes about 12% of the total demand; air pollution control (in wet scrubbers to remove dust and gases) accounts for roughly the same share; and heat transfer (mainly for protecting equipment through extensive water cooling) claims about 75% of the total volume (USDOE, 2013). Specific rate per tonne of crude steel ranges widely among, and within, countries, with a world survey indicating extremes of 1–148 m3/t for input and 1–145 m3/t for discharge. This means that order of magnitude differences are not uncommon even within advanced countries: Germany’s average water consumption in steelmaking declined from more than 35 m3/t in the early 1980s to 10 m3/t in 2010, while the country’s major steel producer averages only 0.77 m3/t of crude steel (Lech Stahlwerke, 2011).
Moreover, care must be taken to compare identical volumes, that is, water actually used by a specific process, water that is repeatedly recycled, and freshwater additions. Gao et al. (2011) provide details for all of these categories for the recent Chinese steelmaking practice and the three rates (in the given order and all in m3/t) are 0.45, 15.47, and 15.97 for coking, 1.75, 68.60, and 70.57 for ironmaking, 0.57, 15.15, and 15.86 for steelmaking, 0.54, 11.86, and 12.40 for continuous casting, and 0.44, 54.70, and 55.63 for hot rolling. These rates (including small volumes for sintering) add up to about 3.8 m3 of consumed water, 167 m3 of recycled water, and 172 m3 of freshly supplied water, with ironmaking accounting for about 40% and hot rolling for more than 30% of freshwater additions. They also put the Chinese national average of consumptive use at 7–8.3 m3/t, with major enterprises averaging 2.5–6.7 m3, the
highest total being, unfortunately, for the Baotou steelworks in the arid north.
The best recent appraisal of average American requirements puts them on the order of 300 m3 (300 t) per tonne of steel, a total that includes new supply as well as water that has been recycled and reused (Worrell et al., 2010). Recycling is practiced widely but inevitable evaporation requires between 50 and 100 m3/t of fresh, make-up water that must be supplied either from public sources or from nearby water bodies. But when Horie et al. (2011) compared water use in steelmaking in Japan, China, and the United States, he used an impossibly low average water footprint (direct and indirect withdrawals) of just 0.62 m3/t of BF/BOF steel in Japan, 0.99 m3/t in China, and about 5 m3 in the United States. In contrast, the latest Polish figures appear quite realistic: direct freshwater use per tonne of crude steel in integrated mills at about 105 m3, with circulating water amounting to about 35 m3 (Burchart-Korol & Kruczek, 2015).
Other published specific rates (all in m3 or t/t of steel) are 5 for sinter plants, 9–10 for EAFs, 12–13 for cold rolling, 13 for continuous casting, 16 for BF, 17 for BOF, 32–40 for various hot-rolling processes, and 37 for coking. Given these substantial differences, quoting a representative global average may be rather imprudent, but the World Steel Association has offered one, based on data collected from 29 steel plants worldwide, and ended up with an average consumption of 28.6 m3/t and discharge of 25.3 m3/t for the integrated steelmaking, and, respectively, 28.1 and 26.5 m3/t for EAFs (WSA, 2011a).
Steel Scrap
Steel does not lose any of its many desirable physical properties by being remelted and recast, and hence it can be recycled (inevitable handling and processing losses aside) ad infinitum. If care is taken to eliminate undesirable elements from steel scrap, then the recycled material can be used to produce casting components and finished products (be they boilers or bolts) of the same quality as those made from primary pig iron—and do so with a fraction of energy required for primary metal production. As a result steel is the world’s most recycled metal. The mass of collected steel is also much larger than for any other recycled material. In the US mass aggregates in 2015 were, with steel equal to 100, about 67 for paper, 6 for aluminum, 4 for glass, and just 3 for plastics (SRI, 2015).
Metal recycling has increased substantially since the 1960s, but the effort still has a long way to go: a UNEP report indicates that only 18 out of 60 metals have end-of-life recycling rates in excess of 50% (fortunately, the list includes not only iron and aluminum but also copper, nickel, manganese, cobalt, and chromium as well the three precious metals, silver, gold, and platinum) and only three rates at between 25% and 50% (UNEP, 2011). Depending on the country, actual end-of-life recycling rates are up to 90% for steel, up to 70% for aluminum, and 60% for nickel but only a bit above 50% for copper.
EAFs are the dominant processor of scrapped ferrous metal that is also charged, in smaller specific amounts, to BOFs. Steel scrap is usually classified into three major streams: home, prompt, and obsolete scrap. Home, or circulating, scrap (also called own arisings) is produced within iron and steel plants and is obviously the easiest material to be recycled, often in a matter of days or weeks after its generation; moreover, its composition is perfectly known, eliminating any quality concerns, and in 2012 it amounted globally to 200 Mt or about 35% of all scrap and the equivalent of 13% of that year’s total steel production (BIR, 2014). The rest of the recycled scrap, globally about 65% of the total in 2012 (370 Mt), is purchased domestically or it is imported, and most of it (about two-thirds) is old metal (obsolete scrap).
Prompt (new) scrap originates from a wide variety of metalworking and manufacturing enterprises that use steel to make metal parts, assemblies, or finished products ranging from white goods (refrigerators, washing machines) to heavy land-moving machinery; composition of this scrap is also well known, but it must be collected and transported to mills and hence it becomes available in a matter of months. Estimates of prompt scrap as a share of finished steel products range between 10% and 15% on the all-industrial basis, with rates as low as 8% in the production of containers and as high as 20% for making some car parts (Kozawa & Tsukihashi, 2011). Obsolete scrap comes from old, discarded metal, some of it years and some of it decades old, with steel often containing undesirable elements in alloys and coatings or commingled with other metals and nonmetallic materials in still-assembled appliances, tools, transportation equipment (ships, railway cars, vehicles), and heavy machinery or embedded in waste from demolished buildings.
Ranges of lifespans of steel-based products differ among countries, but the following numbers are good average (and extreme) approximations (Dahlström et al., 2004; Kozawa & Tsukihashi, 2011; Pauliuk et al., 2013): 60–75 (20–100) years for structural steel in buildings, dams, and bridges; 60 years for oceangoing vessels; 30 (10–40) years for industrial machinery (including turbines and engines); 15 (5–20) years for automobiles and durable metal goods; 10 years for boilers and drums; and as little as 1 or 2 years for cans and metal boxes and 5 years for containers. Rail track life-span is more complicated: before the metal gets recycled, rails can first be reused by swapping the tracks, and later they can be installed on secondary lines with less traffic.
Because the metal is magnetic, recycled steel can be easily separated from nonmagnetic materials, usually in rotating drums after shredding. Screening, pressurized air, and liquid flotation are also used for separation. Admixture of other metals, nonmetallic wastes, and hazardous compounds is a common occurrence with obsolete scrap, whose rusting surfaces can be heavily contaminated by toxic compounds, while the metal itself may contain impermissibly high levels of copper, tin, nickel, chromium, or molybdenum: their concentrations should be generally less than 0.1% for structural steel. Because these metals cannot be separated from the steel, their presence must be diluted by mixing in more high-purity scrap or pig iron. Sized and sorted metal is compacted into small blocks for convenient handling and for land or water-borne transport to domestic steel mills or for international, and increasingly intercontinental, exports.
But in many cases it requires a great deal of effort to get sized, sorted, and compacted blocks of metal ready to be charged into EAFs and also BOFs. Because discarded steel objects come in so many sizes, the recycled material must be first reduced to dimensions suitable for easy handling. The greatest challenge is cutting heavy steel plates of ship hulls and dismantling giant oil tankers, bulk carriers, and cruise liners: some 90% of this difficult work is done (after stripping away the nonmetallic components) by gas and plasma torches at sprawling ship-breaking sites on the seashores and in the shallow waters of Pakistan (Gadani near Karachi), India’s Gujarat (Alang), and Bangladesh (near Chittagong). Despite the Hong Kong International Convention for the Safe and Environmentally Sound Recycling of Ships adopted in 2009 (IMO, 2015), thousands of laborers employed at those sites work commonly without any protective gear, and often barefoot, as they perform series of hazardous tasks while walking (and living) on grossly polluted beaches (Rousmaniere & Raj, 2007).
In contrast, car recycling, the dominant source of obsolete steel in affluent Western countries and in Japan, is highly mechanized. After removing some components that could be resold or recycled (stereo, alternator, some engines), and separating larger nonmetallic parts (some of which, including batteries, tires, and plastics, are also partially or largely reusable) and draining recyclable fluids (gasoline, oil, coolant, transmission, and windshield), car bodies are compacted in a crusher by hydraulic machinery operating under high pressure; the flattened bodies are loaded on trucks or railway cars for transport to mills or for export or are shredded at collection sites (Fig. 6.5). In the United States, a country with a high rate and a long history of car ownership, vehicle recycling is a major industry. Annual totals of scrapped vehicles rose from about 2.5 million units in 1950 to 8.3 million vehicles in 1970, reached a peak of 14.3 million in the year 2000, and has since fluctuated between 11.4 and 14.2 million units (USDOT, 2015).
Figure 6.5 Flattened cars ready to be shredded and charged into EAFs to make new steel. Corbis.
Two countervailing trends affected the total scrapped mass: vehicles got heavier, but the shares of iron and steel declined as aluminum and plastics (and most recently also composite materials) replaced many parts formerly made of cast iron and carbon steel. During the 1950s, ferrous metals made up 94–97% of all metallic components, while their recent shares are down to about 80% of all metals and to between 60% (for hybrids) and 70% for all materials (Field et al., 1994; Jody et al., 2009). On the other hand, cars now contain more high-strength steel: its mass is now more than double that used during the 1970s (for more on this aspect see the next chapter on steel uses).
The United States now has about 12,500 vehicle dismantlers, about 2500 scrap processors, and more than 300 heavy car shredders; the industry employs about 30,000 people, and in 2012 it recycled 14.8 Mt of iron and steel from 11.8 million end-of-life vehicles whose mass is now about 60% iron and steel (Fenton, 2014). In 2014, this automotive scrap trade (including exports) was valued at more than $26 billion, and the total made the US automotive recycling the country’s 16th largest industry. In 2012, about 88% of domestic scrap consumption was destined for iron and steel mills (of which about 80% went to EAFs and 15% was charged to BOFs), and the ferrous castings industry bought most of the rest.
The recycling rate for automotive scrap fluctuates with ups and downs of the car market: in recent years it peaked at 121% in 2009, and it was only 85% in 2013, with the 10-year (2004–2013) average of 103% (SRI, 2014). The recycling rate of US structural steel (beams and plates) has been nearly as high (averaging 98% in recent years); the rate for obsolete appliance scrap consisting mainly of discarded refrigerators, washing machines, dryers, dishwashers, and stoves and containing about 60% of steel, is about 90%. And despite the fact that reinforcement bars in concrete are obviously much more difficult to recycle, their reuse reached also 70% in 2012 compared to just 40% during the late 1990s. Another major scrap stream comes from containers, mostly from steel cans used by many industries for packaging and distributing food, pet food, aerosols, personal care and home care products, lubricants, and paints.