by Vaclav Smil
4
New Materials and New Syntheses
The careful text-books measure
(Let all who build beware!)
The load, the shock, the pressure
Material can bear.
So, when the buckled girder
Lets down the grinding span,
The blame of loss, or murder,
Is laid upon the man.
Not on the Stuff—the Man!
Rudyard Kipling, Hymn of Breaking Strain
Motion and speed—rotation of steam turbines and electric motors, reciprocation of internal combustion engines, travel by new automobiles and bicycles, promise of airplanes—were the obvious markers of the epoch-making pre-WWI advances. At the same time, another kind of less flamboyant innovation was changing the societies of Western Europe and North America. None of those new speedy machines would have been possible without new superior materials whose applications also revolutionized construction. The bold cantilever of the Firth of Forth bridge, shown in the frontispiece to this chapter,
FRONTISPIECE 4. The foreground of this dramatic photograph, taken in April 1889, shows the bottom member of the unfinished North Queensferry cantilever of the Forth Bridge and Garvie main pier in the background. Reproduced from the cover of The Illustrated London News, October 12, 1889.
was thus no less a new departure and an admirable symbol of the 1880s than were Benz’s engines or Tesla’s motors. The bridge was the first structure of its kind to be built largely of steel: only that alloy could make this defiant design possible as only that material could bear the loads put on those massive cantilevers.
This was not the first use of metal in bridge building: wrought-iron chains were used for centuries to suspend walkways across deep chasms in China, and the era of cast iron structures began in 1779 when Abraham Darby III completed the world’s first cast iron bridge across the Severn’s Shropshire Gorge. But these were expensive exceptions as wood continued to rule even the most advanced pre-industrial societies, be it Qianlong’s China or the Enlightenment France. Wood was the dominant material in construction, farm implements, artisanal machinery, and household objects. If these items, and other objects made of natural fibers, were to be excluded from a census of average family possessions, then the total mass of materials that were not derived from biomass amounted to as little as a few kilograms and typically to no more than 10 kg per capita. Moreover, those artifacts were fashioned from low-quality materials, with cast iron pots and a few simple clay and ceramic objects dominating the small mass of such possessions.
In contrast, by the end of the 20th century an average North American family owned directly more than2tofhigh-performance metal alloys in its cars, appliances, furniture, tools, and kitchen items. In addition, two-thirds of these families owned their house, and its construction needed metals (iron, steel, copper, aluminum), plastics, glass, ceramics, paper, and various minerals. Any reader who has never tried to visualize this massive increase of material possessions will enjoy looking at Menzel’s (1994) photographs that picture everything that one family owned in 30 different countries (from Iceland to Mongolia) during the early 1990s. Differences between possessions in some of the world’s richest and poorest countries that are so graphically captured in Menzel’s book are good indicators of the material gain that distinguished modern economies from pre-industrial societies.
And, of course, total per capita mobilization of materials in modern economies, including a large mass of by-products and hidden flows (e.g., coal mining overburden, ore and crop wastes, or agricultural soil erosion), is vastly greater than direct household consumption or possessions. Studies of such resource flows show that annual per capita rates during the mid-1990s added up to 84 t in the United States, 76 t in Germany, and 67 t in the Netherlands (Adriaanse et al. 1997). When the U.S. total is limited only to materials actually used in construction and to industrial minerals, metals, and forestry products, its annual rate shows increase from about 2 t per capita in 1900 to about 10 t per capita after 1970.
This great mobilization of nonbiomass materials began during the two pre-WWI generations. In 1800 steel was a relatively scarce commodity, aluminum was not even known to be an element, and there were no plastics. But by 1870 the world’s annual output of steel still prorated to less than 300 g per capita (yes, grams); there was no mass production of concrete, and aluminum was a rarity used to make jewelry. And then, as with electricity and internal combustion, the great discontinuity took place. By 1913 the world was producing more than 40 kg of steel per capita, a jump of two orders of magnitude in half a century, and the American average was more than 200 kg/person. The cost of aluminum fell by more than 90% compared to 1890, and concrete was a common, although an oft-reviled, presence in the built environment.
But the material revolution did not end with these newly ubiquitous items. Maturing synthetic inorganic chemistry began producing unprecedented volumes of basic chemicals (led by sulfuric acid), and by 1909 Fritz Haber succeeded to do what generations of 19th-century chemists had tried to accomplish as he synthesized ammonia from its elements, an achievement that opened the way for feeding 6 billion people by the end of the 20th century (Smil 2001). On the destructive side of the technical ledger, chemistry moved from gunpowder to nitroglycerine and dynamite and beyond to even more powerful explosives. Dynamism of the Age of Synergy thus did not come only from new energies and new prime movers; it was also fashioned by new materials whose quantity and affordability combined with their unprecedented qualities to set the patterns whose repetition and elaboration still shape our world.
Steel Ascendant
Steel is, of course, an old material, one with a long and intricate history that is so antithetically embodied in elegant shapes and destructive power of the highest quality swords crafted in elaborate ways in such far-flung Old World locations as Damascus and Kyoto (Verhoeven, Pendray, and Dauksch 1998). But it was only during the course of the two pre-WWI generations that steel became inexpensive, its output truly massive, and its use ubiquitous. Unlike in the case of electricity, where fundamental technical inventions created an entire industry de novo, steelmaking was a well-established business long before the 1860s—but one dominated by artisanal skills and hence not suited to high-volume, low-price production that was to become a hallmark of the late 19th century.
Lowthian Bell (1884:435-436), one of the century’s leading metallurgists, stressed that by 1850 “steel was known in commerce in comparatively very limited quantities; and a short time anterior to that period its use was chiefly confined to those purposes, such as engineering tools and cutlery, for which high prices could be paid without inconvenience to the customer.” But by 1850 at least one cause of the relative rarity of steel was removed as the production of pig (cast) iron, its necessary precursor, expanded thanks to taller, and hence much more voluminous and more productive, coke-fueled blast furnaces. Bell concluded that the typical furnaces were too low and too narrow for the efficient reduction of iron ore, and his design increased the overall height by 66%, the top opening by more than 80%, and the hearth diameter by 33% (Bell 1884; figure 4.1).
Other important pre-1860s innovations were the introduction of hot blast (this increased combustion efficiency), the recovery and reuse (for heating of the blast air) of CO-rich hot gases escaping from the furnace’s open top, freeing of the hearth (support of the furnace’s outer casing by cast iron pillars made the hearth accessible from every direction and allowed the placing of a maximum number of tuyeres, or blast-delivering nozzles), and capping of the furnace with cup-and-cone apparatus introduced by George Parry in 1850 (it also enabled a more even distribution of the charge inside the furnace). During the latter half of the 19th century, the technical primacy in ironmaking passed from the British to the Americans (Hogan 1971). Before 1870, the world’s most advanced group of tall blast furnaces operated in England’s Northeast, using Cleveland ores discovered in 1851 (Allen 1981). Pennsylvanian ironmakers then took the lead,
especially with the furnaces built at Carnegie’s Edgar Thomson Works: their hearth areas were more than 50% larger; their blast pressures and rates were twice as high.
FIGURE 4.1. Increasing height and volume of blast furnaces, 1830–1913. Based on data in Bell (1884) and Boylston (1936).
These advances resulted in unprecedented outputs of more than 300 t/day by 1890. But by 1900 nearly all U.S. coke was still made in inefficient beehive ovens rather than in by-product coking batteries, which nearly halved the coking cycle, recovered valuable gases and chemicals, produced higher coke yield, and could use a greater variety of coals (Porter 1924). Abundance of excellent coking coals and cheapness of the beehive ovens are the best explanations of this lag. As a result of these cumulative advances, the global output of pig iron roughly doubled to 10 Mt between 1850 and 1870, and then, as Americans and Germans took the technical lead away from the United Kingdom, it jumped to more than 30 Mt by 1900 and reached 79 Mt by 1913 (Kelly and Fenton 2003; figure 4.2).
This growth was not driven primarily by new technical capabilities that made such outputs of pig iron possible, but by rapidly rising demand for steel whose inexpensive production became a reality thanks to new processes that could convert pig iron into an alloy with superior qualities and hence with a much larger scope of applications than cast iron. Both pig iron and steel are alloys whose composition is dominated by the most common of all metals, but whose differences in carbon content translate into special physical, and hence also structural, properties. Cast iron contains between 2% and 4.3% carbon, while steel has merely 0.05-2% of the element (Bolton 1989).
FIGURE 4.2. Pig iron smelting, 1850-1913. Plotted from data in Campbell (1907) and Kelly and Fenton (2003).
Cast iron has very poor tensile strength (much less than bronze or brass), low impact resistance, and very low ductility. Its only advantage is good strength in compression, and its widest uses have been in such common artifacts as water pipes, motor cylinders, pistons, and manhole covers. In contrast, the best steels have tensile strengths an order of magnitude higher than does pig iron, and they can tolerate impacts more than six times greater. They remain structurally intact at up to 750°C, compared to less than 350°C for cast iron. Addition of other elements—including, singly or in combination, Al, Cr, Co, Mn, Mo, Ni, Ti, V, and W, in amounts ranging from less than 2% to more than 10% of the mass—produces alloys with a variety of desirable properties. Low-carbon sheet steel goes into car bodies; highly tensile and hardened steels go into axles, shafts, connecting rods, and gear. Stainless steels are indispensable for medical devices as well as for chemical and food-processing equipment, and tool steels are made into thousands of devices ranging from chisels to extrusion dies.
Pre-industrial societies were producing limited amounts of steel primarily by a process of cementation, which added desired amount of carbon to practically carbon-free wrought iron. The alloy remained a commodity of restricted supply even by the mid—18th century, when cast iron production began rising substantially. Benjamin Huntsman (1704-1776) began producing his crucible cast steel by carburizing wrought iron during the late 1740s, but that metal was destined only for such specialized, limited-volume applications as razors, cutlery, watch springs, and metal-cutting tools. As the advancing industrialization needed more tensile metal, particularly for the fast growing railways, wrought iron filled the need.
This iron had to be produced in puddling furnaces by the sequence of reheating the brittle pig iron in shallow coal-fired hearths, and pushing and turning it manually with long rods in order to expose it to oxygen and to produce an alloy with a mere trace (0.1%) of carbon. Later this extraordinarily hard labor—which involved manhandling iron chunks of nearly 200 kg for longer than an hour in the proximity of very high heat—was done mechanically. The puddled material was then, after another reheating, rolled or hammered into desired shapes (rails, beams, plates). The first innovation that made large-scale steelmaking possible was introduced independently and concurrently by Henry Bessemer (1813-1898) in England and by William Kelly (1811-1888) in the United States. Bessemer revealed his process publicly in August 1856 and patented it in England (G.B. Patent 2219) in 1856 and in the United States (U.S. Patent 16,083) in 1857 (Bessemer 1905).
Molten pig iron was poured into a large pear-shaped tilting converter lined by siliceous (acid) material, and subsequent blasting of cold air through tuyeres would decarburize the molten metal and drive off impurities. Kelly rushed to file his patent (U.S. Patent 17,628 in 1857) only after he learned that Bessemer filed his in the United Kingdom, but was declared the inventor of the process by U.S. courts. Hogan (1971) compares the key sentences from both patent applications in order to demonstrate how similar was the reasoning behind them. The blowing period lasted as little as 15 and usually less than 30 minutes per batch, and it produced spectacular displays of flames and smoke issuing from the converter’s mouth (figure 4.3). But the process proved to be a great disappointment: while the air forced into molten iron would burn off carbon and silicon, phosphorus and sulfur were left behind.
FIGURE 4.3. Turning gears and a cross section of Bessemer converter pouring the molten metal (top; reproduced from Byrn 1900) and a converter in operation at John Brown & Co. Foundry (bottom; reproduced from The Illustrated London News, July 20, 1889).
This drawback was not discovered during Bessemer’s pioneering work because, by chance, he did his experimental converting with pig iron made from Blaenavon ore that was virtually phosphorus-free. The simplest solution to the phosphorus problem was to use iron ores of great purity, but their supply was obviously limited. A technical fix for the sulfur problem that made the Bessemer process much more widely applicable was discovered by Robert Forester Mushet (1811-1891). This experienced metallurgist added small amounts of spiegel iron, a bright crystalline iron ore with about 8% Mn and 5% C, to the decarbonized iron in order to partially deoxidize the metal (Mn has a great affinity for O2) as well as to combine with some of the sulfur and remove the resulting compounds in slag. Patent specifications for this process were filed in 1856 (G.B. Patent 2219), with most of the rights assigned to the Ebbw Vale Iron Co., which financed Mushet’s experiments (Osborn 1952).
When the company failed to pay a required patent stamp duty, the process became a public property, and it could be freely used by Bessemer in licensing his technique. By 1861 Bessemer steel was being rolled into rails in a number of mills around England, and it was first produced in the United States in 1864. One more essential step was needed to make the process universal: to remove phosphorus from pig iron in order to be able to use many iron ores that contain the element. Many engineers tried for several decades to solve this challenge, and the solution was found by two young metallurgists, Sidney Gilchrist Thomas (1850-1885) and his cousin Percy Carlyle Gilchrist (1851-1935). Their reasoning was not new: to use a basic material that would react with the acidic phosphorus oxides present in the liquid iron and remove them in slag.
Practical realization of the idea had to overcome a number of problems, beginning with the preparation of durable basic linings and ending with the challenge of dumping large volumes of slag (Almond 1981). The cousins persevered, and after several years of experiments with different linings, they produced hard, dense, and durable blocks from impure limestone and sealed the joints with a mixture of tar and burned limestone (or dolomite). As the basic lining would be insufficient to neutralize the phosphorus compounds, they also added lime to the charged ore. In order to safeguard their process, Thomas and Gilchrist took out a dozen British and foreign patents between 1878 and 1879, and their innovation was finally acclaimed at the Iron and Steel Institute meeting in May 1879.
Soon afterward, the production of basic Bessemer steel took off throughout the continental Europe, where poor ores were widely used, and Germans turned the phosphoric slag into a fertilizer by simply grinding it. By 1890, nearly two-thirds of all European steel was smelted by the basic Bessemer process. This technique produced most of the w
orld’s steel between 1870 and 1910. Its share in the American output peaked at 86% of the total in 1890 (Hogan 1971). But this success was short-lived as it was a different procedure, one whose first practical application was one of the markers of the beginning of the Age of Synergy, that proved to be an epoch-making innovation and produced most of the world’s steel for more than two-thirds of the 20th century.
Open-Hearth Furnace
The story of open-hearth steelmaking is one of the best illustrations of difficulties with attributing the origins of technical inventions and dating them accurately. The principal inventor of the furnace was one of four brothers of the German family that has no equals as far as the inventiveness of siblings is concerned. We have already met Werner, the founder of Siemens & Halske Co. and one of the inventors of self-excited dynamos (see figure 2.8). The adaptation of open-hearth furnace for more efficient steelmaking was the idea of Carl Wilhelm, known as William after he became a British subject in 1859 (figure 4.4). Wilhelm, an inventor and promoter in the fields of thermal and electrical engineering, elaborated the idea in cooperation with his younger brother Friedrich (1826-1904).
