by Vaclav Smil
Figure 3.6 Three-wheeled steel riding plow and grain harvester (mechanical reaper) made, respectively, by Deere & Company in Moline, Illinois and D.M. Osborne & Company in Auburn, New York, during the 1880s. VS archive.
Diffusion of steel moldboards was greatly helped by the adoption of sulky plows (riding plows with a steel seat) and gang plows (with two or more moldboards) starting during the late 1860s. Mechanical reapers were introduced in the 1830s—by Obed Hussey (1790–1860) in 1833 and by Robert McCormick (1780–1846) in 1837—but their mass diffusion came only after 1850 (Fig. 3.6). Without these implements it would have been impossible to open North America’s central grasslands (US Great Plains and Canada’s Prairies) for grain cultivation, and to harvest large fields in a timely manner. American agriculture—thanks to its decades-long march westward that created new homesteads, farms, and enclosed pastures—was also the leading consumer of barbed wire made from galvanized steel. Joseph Glidden (1813–1906), an Illinois farmer, introduced the first (and still frequently made) twist design in 1874, and it was followed by hundreds of permutations (McCallum & Frances, 1965).
Grain and hay production further benefited from the expanding use of other steel implements and machines, including harrows, diskers, seeders, hay balers, and horse-drawn, and since 1911 also self-propelled, combines. In aggregate, advancing mechanization, first of North American and later also of Western European agriculture, made the newly redundant labor available for expanding factory employment, and it was a key factor behind rapid urbanization and industrialization, a perfect example of positive-feedback dynamics as more steel for the rural economy led to a higher steel demand in more populous urban settings.
Traditional industries whose expansion and modernization depended on new steel components and machines included textile production (carding, spinning, weaving), paper-making, and glass-making. In 1904 Michael Joseph Owens (1859–1923) introduced the first fully automatic glass-blowing machine able to make 2500 milk or beer bottles per hour. Steel also went into new machine designs for food processing (canning), brewing, and wine-making and many other operations handling liquids and doughs that took advantage of steel tanks and containers. Canning was initially done only in glass, the first metal containers (tin-plated) were produced in the second decade of the nineteenth century, but food in steel cans became common only after 1850.
Steel in energy industries The rapid growth of fossil fuel production created entirely new markets for steel and iron. The energy sector’s new demands for steel were growing especially fast in the United States. This growth is best appreciated by looking at the expansion of the country’s coal and oil production: the former rose nearly sevenfold between 1850 and 1913, the latter (began in 1859) increased twelvefold between 1870 and 1913 (USBC, 1975). Coal mining required drilling, hoisting, and transportation equipment; oil and gas extraction needed drilling rigs and pipes, storage tanks, pipelines, refineries, and gas-processing facilities.
Pipeline construction was made safer thanks to the invention of seamless steel pipes by Reinhard (1856–1922) and Max Mannesmann (1857–1915) in 1885 (Salzgitter Mannesmannröhren-Werke, 2015). Soon afterward came the pilger rolling (reducing the pipe’s diameter and wall thickness while increasing its length) and the combination of the two techniques has been the dominant way of pipe production ever since. And before pipelines became ubiquitous the most commonly used container holding oil was a steel barrel chosen in 1872 by the US Bureau of the Census as the standard measure of crude oil output and trade (it contains 42 US gallons, nearly 160 L). Only a small fraction of global oil trade, carried overwhelmingly by pipelines and tankers, now moves in barrels, but the anachronistic unit is still preferred by the oil industry as a measure of its output.
Another large and entirely new market was created by generation, transmission, and conversion of electricity. In 1882 Thomas Edison (1847–1931) built the first coal-fired power plant in Manhattan, and in the same year America’s first small hydroelectricity station was built in Wisconsin (Smil, 2005). Steel was required for increasingly larger boilers (whose walls were lined with coiled tubing containing circulating water), steam turbogenerators (commercialized for the first time during the late 1880s and now still the world’s most powerful prime movers with the largest unit capacities surpassing 1 GW), water turbines, generator halls (floors to support heavy machinery), and transmission towers and wires (modern high-voltage wires use aluminum with a steel core).
Yet another large market arose due to destructive uses of steel to create and expand modern armaments. First came the introduction of better infantry rifles, then the invention of the self-powered machine gun by Hiram Maxim (1840–1916) in 1884, and the production of steel-encased grenades and bombs and (starting in the 1880s) the construction of large and heavily armored battleships (the famous Dreadnought was built only in 1906). Switching to the opposite end of the scale of objects, and also indicating the breadth of new consumer items whose development took advantage of inexpensive steel, here is a brief list of small steel products invented or introduced between 1850 and 1914 that are still in demand today (Smil, 2005): paper staplers and staples (1868), shower sprinkle heads (1874; alas, too many are now in an even cheaper plastic), Swiss Army knives (1891, now in scores of designs and also in different colors), zippers (clasp lock) in 1893, spring mouse traps (1894, certainly more environmentally friendly than warfarin), and safety razor blades (1904).
And before turning to the use of steel in transportation and construction I should note that the expanded use of the metal necessitated new ways of its examination and quality testing that required many steel-based tools and devices. Steel’s microcrystalline structure could be best examined under microscope (after polishing and etching the samples), and stress tests of the metal were done on machines that could exert more than 450 t in compression and 360 t in tension. But the greatest testing advances came with Wilhelm Roentgen’s (1845–1923) discovery of x-rays in 1895 and with the introduction of x-ray diffraction in 1912 as these techniques allowed, for the first time, nondestructive examination of steel parts. The American Society for Testing Materials (ASTM) was established in 1898.
Steel in Transportation and Construction
Many important parts of traditional personal and freight transportation on land and water were made of iron, ranging (on land) from bits and shoes for horses and springs and wheel rims for coaches and wagons, to (on water) nails, brackets, belaying pins, chains, and anchors (and the all-important magnetic compass!). As important as they were functionally, in mass terms these iron components remained relatively minor accessories to the overwhelmingly wooden transportation equipment of antiquity, the Middle Ages, and of the early modern era. Only the confluence of the widespread adoption of the steam engine, the first mechanical prime mover that could power vehicles and vessels, and the availability of inexpensive Bessemer steel created new markets for the metal that soon resulted in production orders of magnitude higher than during the pre-1850 era.
The development of the steam engine for transportation became possible only after the expiry of Watt’s original patent in 1800. Watt was afraid to work with high-pressure boilers, but only they could be made (relatively) light enough for mobile uses. Steam-powered vessels came first—Patrick Miller’s (1731–1815) Charlotte Dundas in 1802 in England, Robert Fulton’s (1765–1815) Clermont in 1807 in the United States, and the first Atlantic crossing by Royal William in 1833 (Fry, 1896). On land, Richard Trevithick’s (1771–1833) machines on cast iron rails (in 1804) were premature, and the first scheduled railway service between Liverpool and Manchester began only in 1830. Both of these markets multiplied the steel demand for the remainder of the nineteenth century: by that time the expansion of railways in Europe and North America slowed down, but more steel was needed for ocean-going vessels (including new massive war ships), and the first decade of the twentieth century brought new demand for steel created by the world’s first mass-produced car, Ford’s Model T introduced in 1
908.
Steel for railways Rapid expansion of European and North American railways added up to the largest market during the first decades of inexpensive steel. The railroad era began in 1830 with the world’s first 56-km-long intercity link between Liverpool and Manchester. By 1900 the United Kingdom had about 30,000 km of railways, Europe had approached 250,000 km, Russia had 53,000 km, the total length of US railways had surpassed 190,000 km (including three transcontinental routes), and the global aggregate was 775,000 km (Williams, 2006). All rails laid down before the late 1850s were made from low-carbon wrought iron produced by puddling, and a significant amount of this metal for the US rails was imported from the United Kingdom even after the Civil War (and, in turn, the United Kingdom still imported Swedish wrought iron during the 1850s), and US production of wrought iron rails peaked only in 1872 (Birch, 1968).
These rails did not oxidize but they could not withstand the shocks of moving trains for a long time and had to be replaced as early as every 6–12, and later every 36, months, while steel rails would last a decade. Axle breakages were common, with the one in 1842 causing France’s first great railroad disaster at Meudon where 55 passengers were burned alive (Caron, 2013). Faulty tracks on the early railways were one of the major reasons for the development and institutionalization of metal testing and standardization. Bessemer steel improved the quality of tracks as rails became its first market to conquer. The first steel rails were laid in England in 1861, and by the mid-1870s they dominated the British market, were exported to the United States, and allowed the unprecedented rail expansion on three continents (Europe, Asia and North America). In the United States iron rails still accounted for about 70% of all tracks in 1880, but two decades later their share was below 8%, and open-hearth steel was dominant everywhere.
Early rails weighed 20–25 kg/m, by 1880 the mean was about 30 kg/m and maxima reached 45–50 kg/m during the 1890s (Hogan, 1971). Using a conservative mean of 30 kg/m—to get an estimate of the right order of magnitude—would imply that between 1850 and 1900 the construction of railways consumed at least 20 Mt (and perhaps as much as 25 Mt) of iron and steel for the original tracks (Smil, 2013), and at least twice, and perhaps up to three times, as much metal for their replacement after they were in service for no more than 10 years (Ransom, 1989). For comparison, the nineteenth-century railway expansion also consumed at least 160 Mt of sawn wood for sleepers (this total included the initial emplacement and periodic track renewal) and I have also estimated—using a conservative assumption of at least 2000 t of ballast (crushed stones packed underneath and around ties) per kilometer—that some 2 Gt of coarse gravel were needed to support the tracks (Smil, 2013).
Steel also became an increasingly important component of tunnel construction, and (together with cast iron) it was an indispensable material for every other key component of expanding rail transportation, ranging from locomotives, coal and water tenders, and passenger and freight cars to switching and signaling equipment, wires, fencing, walkways, elevated crossing, and large railway terminals (Fig. 3.7). In nineteenth-century Europe the new terminals were typically lofty, open-ended structures covered with glassed roofs, a new style of architecture so memorably depicted in a series of famous 1877 impressionistic paintings of Gare Saint-Lazare by Claude Monet (1840–1926).
Figure 3.7 American steam locomotive: Central Pacific’s model 229, first built in Sacramento in 1882. VS archive.
Steel in shipping Shipping was the other major transportation market for steel. Small iron barges were the first metal-hull vessels built during the second decade of the nineteenth century. In 1833 the Lloyd’s Register approved Sirius, the first iron-hulled steamer; in 1832 Macgregor Laird (1808–1861) designed Alburkah, an iron paddle steamer that he took on a trip to West Africa; and in 1840 Isambard Kingdom Brunel’s (1806–1859) Great Britain was the first iron vessel to cross the Atlantic (Dumpleton & Miller, 1974). The age of steel ships began slowly during the 1860s (helped by the introduction of screw propellers in 1862) and they became common during the 1870s, even before 1877 when Lloyd’s Register of Shipping accepted the metal as an insurable material for ship construction. Concord Line’s Servia, launched in 1881, was the first large trans-Atlantic liner built of steel rather than iron, but it was in service only until 1902 (Babcock, 1931).
The demise of wooden ships took place within a single generation (no sizeable wooden vessels were built by 1900), and the race was on to build ever larger passenger steamers. By 1900 British, German, and French shipyards were building passenger ships that incorporated more than 10,000 t of steel. These efforts peaked with the construction of the largest and the fastest pre–WW I steamships, Lusitania and Mauritania, by the Cunard Line in 1907 (both 31,000 t of gross weight), and the White Star Line’s Olympic, Titanic, and Britannic; each one of those three had gross weights of 46,000 t with hulls of plain carbon steel fastened with wrought-iron rivets (White Star Line, 2008). Titanic became the most famous of all the great liners after it sank on April 12, 1912, following its collision with an iceberg three to six times its mass (Lord, 1955). The quality of the steel used to make the ship’s hull was among the factors suspected to have a role in the catastrophe. After the wreckage was found, remotely controlled vehicles brought up numerous artifacts as well as pieces of steel from the hull that was subject to chemical, micrographic, and mechanical analyses.
Felkins, Leighly, and Jankovic (1998) found that the metal used in the Titanic’s construction was probably the best plain carbon steel available in 1911 but nearly a century later would be unacceptable for any construction projects, particularly not for an ocean liner. The steel was made in acid-lined, rather than basic, open-hearth furnaces in Glasgow, and it had relatively high levels of phosphorus and sulfur (its manganese:sulfur ratio was very low, less than half that in modern steel), which tended to embrittle the metal at low temperatures and hence make it unsuitable for service in cold water (at the time of the collision the ocean near Newfoundland was only –2 °C). While it is impossible to say how much damage would have been suffered if the ship were built with the best modern steel it is most likely that, without the collision, the Titanic could have served, as the Olympic had done, for more than two decades.
Steel in the automobile industry An entirely new transportation sector was created by the introduction of automobiles during the 1880s. Designs of the earliest German models by Carl Benz (1844–1929), Wilhelm Maybach (1846–1929), and Gottlieb Daimler (1834–1900) looked like horseless carriages, with wooden bodies and steel used sparingly for a few key parts (Smil, 2005). Automobiles remained expensive because their production was entirely artisanal (small hand-made series) during the 1890s. In 1901 Mercedes 35-hp, generally considered the first modern car, had its front-mounted four-cylinder engine bolted to a pressed-steel frame, but it, too, was an expensive artisanal product made for rich buyers (Daimler, 2015). Large-scale production of affordable cars began only with Ford’s famous Model T in 1908, the first car made of vanadium steel (Fig. 3.8).
Figure 3.8 Ford’s Model T introduced in 1908 and discontinued in 1927. VS archive.
Properties of this alloy were first identified and studied by J. Kent Smith starting in 1900, and after he became a consultant for Ford’s company the automaker employed United Steel Company in Canton, OH, to produce the first batch of the metal in 1906, and the steel, affordable and easy to machine, eventually constituted half of the mass of the Model T (Misa, 1995). A key part of Ford’s advertisement was to stress the quality of the metal’s superiority:
The Model T is built entirely of the best materials obtainable. No car at $5000 has higher grade, for none better can be bought. Heat treated Ford Vanadium steel throughout; in axles, shafts, springs, gears—in fact a vanadium steel car—is one evidence of superiority.
Nobody disputes that Vanadium steel is the finest automobile steel obtainable …We defy any man to break a Ford Vanadium steel shaft or spring or axle with any test less than 50% more rigid than would be required
to put any other steel in the junk pile …. (Ford, 1909)
As a result, the steel market for the US auto industry expanded from just a few thousand tonnes in 1900 to more than 70,000 t in 1910 (Hogan, 1971). The auto industry’s demand for quality steel also led to the first efforts to standardize the alloys and to specify their permissible composition. In 1912 the Society of Automotive Engineers (SAE) chose descriptive numbers to identify 15 specific classes of steels (e.g., 10- for carbon steels, 23- for nickel steels, 61- for chromium-vanadium steels), and suffixes to indicate the share of carbon (hence 10–60 was plain carbon steel with 0.6% C, 61–25 contained 0.25% C, 0.9% Cr, and 0.18% V). In addition, the SAE also listed two key physical attributes, steel’s elastic limits and toughness, that were expressed, respectively, as percentages of elongation and of reduction in cross-sectional area (Misa, 1995).
Steel in construction The first, and highly visible, use of steel in construction was for new bridges, at first for relatively short crossings, later for unprecedented spans. Shorter bridges used prefabricated iron trellis girders assembled with rivets, and new suspension bridges began to use steel cables instead of wrought-iron chains (Cossons & Trinder, 1979). Perhaps the most iconic pioneering long-span bridge crossed the Firth of Forth in Scotland by a cantilevered 2529-m-long structure designed by John Fowler (1817–1898) and Benjamin Baker (1840–1907), built between 1883 and 1890 and used by the North British Railway (Fig. 3.9). The bridge required 55,000 t of the metal, most of it for its two 104-m- tall towers and massive cantilevered arms (Forth Bridges, 2013), and it has been both admired and derided.
