Still the Iron Age

by Vaclav Smil

  Better steels for appliances now include sheets pre-painted at steelworks (eliminating degreasing and painting by appliance manufacturers) and environmentally friendly products including chromate-free electro-galvanized steels and lead-free alloy-plated sheets (Kubo et al., 2012). In rich countries new purchases are overwhelmingly just for replacement units, while most modernizing countries have a long way to go before their ownership of basic appliances will become saturated. Refrigerators and room air conditioning units have now diffused widely among richer urbanites even in some still very poor countries (big cities in India being the prime example), but ownership of all major appliances remains low in rural regions of the Indian subcontinent, and it is virtually absent in the countryside of sub-Saharan Africa.

  Most people do not think of steel and electronics as a common combination—but 47% of the mass of flat-screen TVs is steel, and stainless steel is an important component of electronic devices and components ranging from desktop computers, printers, and hard disk drives to transformers, cables, and screws. And although there is very little steel in the now ubiquitous cellphones—whose mass is dominated by plastics, special glass, and small amounts of many precious and toxic metals (Cd, Pb, Hg, As, Ni, Ag, Au)—stainless steel is used for parts that require corrosion resistance and strength, such as springs, hinges, and screws (ISSF, 2015).

  Chapter 9

  Looking Back

  Advances, Flows and Stocks

  Abstract

  Since the beginning of the twenty-first century the notion of innovation and technical progress has been, both in the public perception and in the majority of technical writings, conflated with the advances of electronics and it is still expanding applications in communication, business, entertainment, and scientific research. Because of this collective infatuation it is even more important to stress the fundamentals on which this electronic edifice rests: without the constantly modernizing steel industry it would have been impossible to develop modern high-energy civilization where an unprecedented share of the global population enjoys excellent quality of life, where the number of people living in poverty in low-income countries was cut by more than half since 1990, and where there is a scientific, technical, and organizational potential to extend these benefits to additional billions.

  Keywords

  Iron smelting; modern steel; steelmaking transition; industry’s state; flows and consumption rates; steel stocks

  Since the beginning of the twenty-first century the notion of innovation and technical progress has been, both in the public perception and in the majority of technical writings, conflated with the advances of electronics and its still expanding applications in communication, business, entertainment, and scientific research. Because of this collective infatuation it is even more important to stress the fundamentals on which this electronic edifice rests: without the constantly modernizing steel industry it would have been impossible to develop modern high-energy civilization where an unprecedented share of the global population enjoys excellent quality of life, where the number of people living in poverty in low-income countries was cut by more than half since 1990, and where there is a scientific, technical, and organizational potential to extend these benefits to additional billions.

  How has this been accomplished? Iron smelting has a fascinating history longer than 3000 years, and I noted many of its fundamental developments in the first two chapters of this book. Steel, too, has a long history, but the requirements of its production kept its overall mass quite limited until after 1860: the history of large-scale commercial production of steel is thus only about 150 years old when starting the count during the late 1860s with the spreading adoption of Henry Bessemer converters in the United Kingdom. But this century and a half brought many remarkable technical advances that transformed ironmaking and steelmaking into indispensable (albeit now so curiously overlooked and unappreciated) industries supplying a variety of alloys, without whose mass-scale applications there could be no affluence for the haves and no hope for the have-nots.

  No less importantly, more efficient use of raw materials and considerable reduction of specific energy consumption have been integral parts of these production advances and, in turn, they have greatly lowered the industry’s environmental impact: producing today’s mass of iron and steel with the techniques that prevailed a century ago would be materially wasteful, energetically unaffordable, and environmentally intolerable. I will review all of these technical advances by looking at key performance variables in long-term perspectives, summarizing concisely many trends that I followed in greater detail in the first seven chapters of the book and depicting them in graphs.

  Afterwards I will turn to do a brief assessment of the world’s steel industry during the second decade of the twenty-first century by focusing on its many contradictions: overlooked yet indispensable; successful yet imperiled; efficient yet still perceived as environmentally offensive; innovative but fundamentally dependent on aging techniques; able to supply the demand but burdened by excessive capacity; no longer a mass employer but still the basis of productive sectors that provide good wages for millions. This will be followed by a brief quantitative recapitulation of key consumption trends (absolute and relative, in per capita terms) of the past 150 years and by appraisals of their relation to economic development and quality of life.

  Continuing with historical perspectives, I will close the chapter by concentrating on the most obvious aggregate physical manifestation of the world’s massive production and use of steel, namely on the rising national and global stocks of the metal, including their origins and the rates of their accumulation. Only concrete (most of it actually reinforced with steel) offers a similar example of accumulation of anthropogenic materials on scales unprecedented in the history of civilization. Annual additions to global concrete stocks are an order of magnitude larger than the accumulation of in-use steel stocks, but steel stocks are incomparably more important: concrete is rather difficult and costly to recycle, while steel stocks have come to form a new, valuable anthropogenic resource that already supplies raw material for nearly a third of global steel production and whose dependence will only grow, not only because of steel scrap’s rising availability but also because of the numerous environmental benefits of its reuse.

  A Century and a Half of Modern Steel

  Foundations of modern steelmaking were laid by the long development and technical advances of ironmaking. Primary production of iron in blast furnaces is a perfect example of a truly medieval technique (its Western European origins date to more than 600 or 700 years ago) whose principle remains the same but whose size, degree of sophistication (both in terms of design and operation), and resulting gains in productivity have enabled vastly expanded output with reduced use of inputs and with significantly lower costs. Given their fundamental importance in producing most of the civilization’s dominant metal, it is no exaggeration to rank these massive assemblies among the most remarkable artifacts of modern societies (Geerdes, Toxopeus, & van der Vliet, 2009; Peacey & Davenport, 1979; Wakelin & Fruehan, 1999; Walker, 1985).

  By 1850 the best British blast furnaces designed by Lowthian Bell were fueled with coke and received hot blast from larger stoves, and their increasing volume brought higher productivity and lower unit costs of pig iron. And yet these proto-modern furnaces were just toys compared to what was to come. First, the furnaces grew taller and acquired wider hearths: by 1930 the largest ones were twice as tall and had hearth diameters nearly twice as large as in 1830; then they grew stouter: their height has increased only slightly, but by 2015 the diameters of the bellies and hearths of the largest furnaces were twice as large as in 1930 (Fig. 9.1). Doubling of hearth diameters results in quadrupling of hearth areas. Hearth areas increased from only about 2 m2 in 1800 to 20 m2 by 1910 and 70 m2 by 1950, and in 2015 the largest furnaces had hearths on the order of 200 m2 compared to less than 15 m2 at the beginning of the twentieth century.

  Figure 9.1 Changing designs of blast
furnaces, 1830–2015. Based on data in Bell (1884), Boylston (1936), King (1946), Sugawara et al. (1976), and Haga (2004).

  Larger hearths and taller stacks resulted in 24-fold growth of maximum inner volume between 1840 and 2015: Bell’s 1840 blast furnace redesign had about 250 m3 (compared to just 50 m3 in 1810), by 1880 volumes of the largest furnace surpassed 500 m3, the 1500 m3 mark was reached by 1950, and the largest furnaces in 2015, with inner volumes of 5500–6000 m3, were an order of magnitude more voluminous than their predecessors 100 years ago (Fig. 9.2). This dimensional growth has resulted in higher nominal capacities and rising daily productivities. The very first coke-fueled blast furnace, in 1709 in Coalbrookdale, produced just two tonnes of hot metal a day.

  Figure 9.2 Growth of blast furnace internal volumes. Plotted from numerous sources cited in the text.

  With the introduction of hot blast the rates surpassed 50 t/day during the 1840s, reached more than 400 t/day by the beginning of the twentieth century, approached 1000 t/day before WW II, and rose to 10,000 t/day in the mid-1970s, and the largest furnaces now produce around 15,000 t/day (and the record rate for POSCO’s Pohang 4 is about 17,000 t/day), and their design capacities are close to, at, or even above 5 Mt/year, 10 times the best performances achieved immediately after WW II (Fig. 9.3). As explained in some detail in Chapters 6 and 7, these productivity advances were accompanied by declines in specific use of coke and in overall energy requirements. The earliest coke rates, during the mid-1700s (as much as 9000 kg/t of hot metal) were extremely wasteful; by the beginning of the twentieth century typical performances were 1000–1100 kg/t, and by 1950 good rates were just above 800 kg/t.

  Figure 9.3 Growth of daily blast furnace production. Plotted from numerous sources cited in the text.

  Subsequent coke rate reductions resulted from the combination of higher efficiencies of use and partial replacement by other reductants, first by oil and later by natural gas and, above all, by pulverized coal injections. By 1960 the best coke rates were well below 700 kg/t, and since the 1980s the overall consumption of reducing agents has stabilized at around 500 kg/t of hot metal, but by 2010 nationwide coke rates were about 370 kg/t in Japan and less than 340 kg/t in Germany (Lüngen, 2013). Combined with other efficiency gains (see Chapter 7), the energy cost of ironmaking has seen spectacular reductions: the earliest coke-fueled process consumed as much as 275 GJ/t in 1750; by 1900 the best rates were down to about 55 GJ/t, they were just above 30 GJ/t in 1950, and they were mostly between 12 and 15 GJ/t by 2010 (Fig. 9.4).

  Figure 9.4 Energy cost of ironmaking, 1750–2015. Based on Smil (2008) and on post-2000 rates cited in the text.

  Large-scale commercial production of steel began with Bessemer converters; they were soon largely replaced by open hearth furnaces, and this nineteenth-century process dominated global steel output until after WW II, when it was transformed by the global adoption of basic oxygen furnaces and by the rising importance of scrap-based steelmaking in electric arc furnaces. One thing all of these steelmaking techniques have had in common has been the rising product yield (Takamatsu et al., 2014). When measured as a share of initial inputs, early Bessemer converters turned less than 60% of iron into steel, but eventually their yield was above 70%; early open hearths performed no better than early Bessemer converters, but a small number of remaining units now have yields of about 80%. BOFs started much higher, turning about 80% of charged iron into steel in 1952, but now that share is as high as 95%, with EAFs doing even better, with yields rising from about 85% before WW II to 97% today.

  As explained in Chapter 5, major steel-producing countries differed in the onset and tempo of adopting oxygen furnaces, and the share of electric steel has been determined by the domestic availability, or the ability to import, scrap, but by 2013 those two steelmaking techniques accounted for 99.5% of total production: the tiny remainder came from technically ancient open hearth furnaces that still produced about 20% of Ukrainian steel. Globally, oxygen steelmaking went from less than half of the total in 1970 to 72% by 2013, and electric steelmaking rose from about 15% in 1970 to nearly 30% by 2015. Fig. 9.5 illustrates the complete steelmaking transition (from Bessemer convertors to OHF, BOF, and electric arc furnaces) for the United States.

  Figure 9.5 Transitions from Bessemer converters to open hearth furnaces to basic oxygen and electric arc furnaces. Plotted from data in Campbell (1907), Temin (1964), and WSA (2014).

  Improvements in the design and operation of electric arc furnaces increased their size and productivity as they cut their tap-to-tap times, electricity demand, and electrode consumption. By 2010 the shortest tap-to-tap times of about 30 min were just 1/6 of the mean in 1950, and during the same time electricity consumption for the most efficient furnaces fell by half, and electrode carbon used per unit of steel declined from 6 kg/t to just above 1 kg/t (Fig. 9.6). The last fundamental technical revolution in steelmaking was ushered in during the early 1950s by continuous casting. A classical sequence of adoption—initial slow phase (just over 5% of the total production by 1970) was followed by a takeoff (30% by 1980) and market dominance (60% by 1990), and by 2013 98.8% of the world’s steel came out of continuous casting machines.

  Figure 9.6 Advances in operation of electric arc furnaces. Based on Lee and Sohn (2014).

  At the same time, the industry has become an impressively cleaner enterprise. Iron- and steelmaking were traditionally among the most prominent pollution-generating activities of the early and mature industrial era, a combined function of its high-energy intensity and of its lack of pollution controls. But the industry has done well during the post-1960 quest for lower energy use and reduced emissions (McKinsey, 2013; USEPA, 2012). Between 1960 and 2010 specific rates (measured per tonne of hot metal) declined by more than 40% for the overall energy consumption, by nearly 50% for CO2 emissions (measured as weighted average of BF, BOF, and EAF production), and by 98% for dust emissions, while the accident rate was cut by about 90%.

  In 1850, before the modern steel production began, less than 100,000 t of the metal were produced annually, prorating to only about 75 g a year per capita. In 1900 global steel production reached 30 Mt, or about 18 kg/capita, while in the year 2000 the global output of 850 Mt by 2000 t translated into average annual output of 140 kg/capita, nearly 2000 times the 1850 rate. And by 2015 the rate rose further to about 225 kg/capita. Estimates of the pre-1950 world economic product are unreliable, but less questionable totals for the years 1950 and 2000 (expressed in constant $1990) indicate that the steel intensity of the global economic product was more than halved, from about 45 kg to 20 kg/$1,000, but (due to China’s rapid increase of demand) by 2015 the rate was about 10% higher.

  Industry’s State

  Perspectives and judgments keep changing. At the beginning of the twentieth century no product was more fundamental to the rise of America’s economic might than steel, the reality exemplified by the industry’s dominance of the global market. But judging by the tenor of today’s writings on technical advances and on our material world, by the beginning of the twenty-first century steel appears to be rather a démodé product not worthy of any close attention in the new e-world that, as the consensus has it, is dominated by silicon and lithium, by microchips, computers, portable electronics, and batteries, the New Economy that generates and accelerates its own growth.

  Confirmations of these perceptions are easy to find. US Steel, the company established by Andrew Carnegie in 1901 and one of the drivers of America’s twentieth-century economic dominance, was deleted from Dow Jones Industrials in May 1991, followed by Bethlehem Steel in 1997 (Dogs of the Dow, 2015). In the year 2000 the combined market capitalization of America’s 10 largest steel companies had equaled just 10% of the value of Home Depot, a retail company added that year to the list of Dow industrials. For a few months during 2009 US Steel’s market capitalization was below its liabilities to retiree and life insurance plans (O’Hara, 2014), on July 2, 2014, the company was removed from the S&P 500 index, and in Apri
l 2015 its market value was less than $3.5 billion, about 1.5% of what Facebook, a company that has never made anything in its short life, was worth.

  And yet such economic realities are an extremely poor indicator of actual, and truly existential, importance of specific products and activities, and such irrational valuations only reflect speculative biases of investors. After all, it requires only a simple thought experiment to see how misinformed such impressions, and how unrealistic such valuations, are. Obviously, it is entirely possible to have an accomplished, advanced, and affluent civilization without any massive deployment of microchips and portable electronics—indeed, we had such a world until, respectively, the 1960s, when Intel began marketing the first microprocessor, and the 1980s, with the advent of mass ownership of personal computers and the introduction of the first cellphones. But no accomplished, advanced, and affluent civilization would be possible without mass-scale production and ubiquitous use of steel, and no investor driving up stock prices of e-companies could make it through a single day without steel.

  To fulfill this critical role in the world of more than 7 billion people and in the interdependent world economy that is now worth annually more than $100 trillion, the steel industry must be large, and its enormous scale is easily demonstrated by a few global comparisons. Mining of iron ore, now surpassing 3 Gt/year, is by far the largest extraction of the metallic element, while combined primary and secondary steel production of more than 1.6 Gt is about 19 times larger than the aggregate annual output of five other leading metals, with aluminum at nearly 50 Mt, copper at close to 20 Mt, zinc at above 13 Mt, lead at nearly 6 Mt, and tin at less than 300,000 t (USGS, 2014). In addition, about 12% of total coal extraction (roughly 1 Gt) is converted to metallurgical coke, and the industry consumes every year close to 400 Mt of fluxing materials (lime, limestone, dolomite).