Alternative ways of making steel : Retrospective and prospective

J.-P. Birat (Arcelor)

The Steel Industry today is a large and capitalintensive industry, which has reached a high level of sophistication, complexity and efficiency. This benefits the consumer, as the price of steel has been steadily decreasing, which helps bring down the price of consumer goods. Steelmaking has therefore become focused on a small number of process routes, which connect together standard, mainstream and high performance technologies. It seems therefore that the industry is very mature and that little leeway is available for change, as past progress has already brought it close to physical limits. This is a view of the Steel Industry, which is contemporary and central to the strategy of this business. A Steel Mill that does not operate within the paradigms of the mainstream state-of-the art technologies will be doomed by economic troubles and eventual bankruptcy. However, Steel or Iron have remained at the core of civilization for roughly 3,000 years and the model of the Steel Industry discussed above is only roughly 30 years old. There have been many more different models in the past, from Prehistory to History, and it is likely that even more will emerge in the Future. Indeed, it is reasonable to assume that Iron and Steel will continue to be used for a very long time. The present paper discusses Steel from the standpoint of an Enduring Material and the Steel Industry from that of a Cumulative Technology and explores how these two important concepts have led to industry changes in the Past and will most likely contribute to more changes in the Future. From an historical standpoint, iron was initially produced from meteorites, but very quickly moved on to the reduction of ores.

The reduction technology went from a solid-state route to a molten state one and, because carbon was used as the reducing agent, pig iron was invented in addition to the pure iron which was commonplace before. Steel was the next concept, also made initially in the solid state and then in the liquid state. The modern world brought recycling into the picture, because of the sheer size of recycled streams. From an economical standpoint, the drivers for changes were mainly the demand of the economy : From weapons, pots and pans to machines and ubiquitous consumer goods. Looking at a shorter timescale, i.e. the end of the 20th century, other drivers were instrumental in bringing change, such as meeting the steady drop in the market price of steel, improving quality and helping steel be better integrated and used in products (steel solutions or customer engineering). The progress thus generated made it possibly to become and remain environmentally friendly. In the future, what will be the drivers for change ? Mass production at an even larger scale, to meet the growth of the world population and the equalization of the standard of living, in spite of a dematerialization of the economy, will be a major driver. On the other hand, Global Environmental Issues, the first of which will certainly be Global Warming, will reshuffle priorities and call for breakthrough solutions meeting demands different from today's. From a technological standpoint, if one projects far enough in the future, many new possibilities open up, in terms of energy, processes, raw materials and of course, also of new materials competing with steel.

La Revue de Mtallurgie 2004.

La Revue de Mtallurgie-CIT

Novembre 2004

937

INTRODUCTION

Like the statue of Hercules used by Aristotle to explain his theory of causes (1) the role that iron plays in our society and our civilization has many causes. Some are contemporary, modern and even post-modern, while others are fundamental, basic, archetypical and as old as the world or as the history of mankind. The value of Iron and of Steel, not just to the shareholder of the Steel Companies or to their customers but to mankind, is related to the complex and ambivalent relationships that iron entertains with the physical world and the living world, with history and modern society. There is a tension between its enduring and its innovative features, i.e. between the cosmological, geological and historical past on the one hand and the present and the future on the other hand. Between eons and time-to-market or time-to-volume! Between culture or heritage and change and breakthroughs This is what we are planning to review in this article for this jubilee issue of the Revue de Mtallurgie. s

introduce it into an amino-acid or to other molecules, thanks to a specific Fe-O interaction (4). Iron is active in the synthesis of DNA, in the scavenging of free radicals, in the metabolic mechanism that releases energy in the cell by using glucose (2). The prevalent biological redox states of iron are Fe2+ (ferrous), Fe3+ (ferric) and Fe4+ (ferryl). The easy cycling between Fe2+ and Fe3+ makes iron-bearing proteins a biological tool for manipulating individual electrons. Surplus iron is stored in the liver in proteins such as ferritin (several thousands of Fe atoms) or hemosiderin. Bone marrow, where hemoglobin is synthesized, is also rich in iron. A protein called transferrin transfers iron between cells (5) (fig. 1). Iron is also important in the functioning of the brain, of the ear, of the pancreas, etc. There is a competition for iron in the body between the host and bacteria during an infection and the production of transferrin is then accelerated, thus providing a kind of antibiotic effect. Deficiency in iron leads to anemia and a daily intake is necessary, at the level of 7 g for a man and 11 g for a woman (2) with a yield of roughly 25 % compared to the food content. The diet in developed countries easily provides the needs of the body, but anemia is prevalent among several million people elsewhere. Iron is therefore central to life on Earth and probably to life in the Universe, if this is indeed an option. Life is also part of ecosystems and iron plays an important role in controlling their equilibrium. For example, recent experiments have shown that the iron (and manganese) content of the sea in its surface layers is directly related to the teeming of life there, through the trophic chain that connects plankton and fish. Indeed, iron is an essential nutrient of plankton, which itself can feed large fish populations. The issue has been seen as a possible solution path, both for increasing the fish resource and for controlling CO2 emissions from the ocean, as plankton interacts with this gas through photosynthesis (6 - 9). The matter of controlling Global Warming by using this lever, what is called "climate engineering", is far from settled, however.

IRON AND LIFE

Iron has been incorporated early into the fabric of life and its role is essential in all living things, from microorganisms to the complex systems that mammals and men have become. It is a trace element, as there are approximately 4 g of Fe in the human body, with a concentration of 415 ppm in the blood, 3-380 ppm in bones and 180 ppm in tissue (2, 3). Biochemists establish a clear relationship between this essential role of iron and its abundance in the Universe (cf. next section). Iron is present in the blood of vertebrates, as the heme of the hemoglobin of the red blood cells (C34H32FeN4O4), which transports oxygen and CO2 between the lungs and the cells. Iron is also present in various enzymes, called non-heme enzymes, which also act in the exchange of oxygen within the cell by helping free one O atom from an O2 molecule and

Fig. 1 Ferritin and transferrin molecules, the basic proteins that store and transfer Fe atoms in the human body.
938 La Revue de Mtallurgie-CIT Novembre 2004

LE CENTENAIRE

or fusion. A massive star will thus have an onion structure with an iron core and layers of Si, O, C, He and H. It is interesting to note that astronomers measure the age of stars by the ratio Fe/H (11). From the standpoint of elements, i.e. of ordinary matter thus excluding black matter and quintessence (12), the one that are most abundant in the Universe range from hydrogen to iron in the periodic table (13) (fig. 2). Heavier elements, which are synthesized by less likely nuclear reactions, are much fewer. The over-abundance of hydrogen and helium is due to the fact that they were produced in the primordial nucleosynthesis of the big bang. The "onion" star may not be the end of the stellar adventure though, if the mass is big enough, as the increase of the core iron mass will lead to an electron-proton fusion (-reaction), which transforms Fig. 2 Distribution of chemical elements in the Universe. protons into neutrons : the star collapses on itself and turns into a supernova, the ashes of which become a neutron star on the one hand and a stream of inters ONCE UPON A TIME IN stellar matter projected into space on the other hand Further THE UNIVERSE down this path may even lie a black hole! Iron, as a chemical element, was forged by nucleosynthesis in the nuclear furnaces of the stars. Stars are nuclear fusion reactors, where nuclei are formed because the density of matter is large enough for nuclear reactions to take place by nucleon addition or fusion of nuclei. The theory of nucleosynthesis is too complex to be explained here, but the trend is for heavier elements to be formed as the star ages (it heats up due to gravitational densification and reaches a temperature, expressed in millions of K, that depends on its mass). Stars will go through a sequence, which comprises the generation of helium (the present state of the Sun), then of carbon, of silicon and possibly of iron, through the reaction : [(Si28 + He4) + He4) + He4] Fe +
56

New stars will be born from that projected material as well as their pageant of planets. Our own solar system originated in this way and this is the reason why iron constitutes the major part of the heavy telluric planets, to which our Earth belongs. Figuratively speaking, one may say that the first steelmakers operated billions of years ago (more than 4.6 billions, the age of the Solar System) in the core of the stars and produced all the stock of iron atoms from which we are sampling a minute fraction for our own needs today. The Steel "industry" was then quite different from today's ! Metallic iron is the most abundant constituent of planet Earth (3) followed by oxygen, silicon, magnesium, nickel, sulphur, etc. (fig. 3). This stems directly from the features of nucleosynthesis. This iron is in a liquid state mainly, alloyed with nickel in the so-called Ni-Fe core (80 % Fe, 5 % Ni, + Si, O,

Small stars will stop their evolution on the way, depending on their mass. It takes many times the mass of the Sun to enter the club of iron-producing stars and an eventual temperature larger than 3 x 109 K ! (10). Iron Fe56 is actually the heavier element that can be generated in this manner, because this peculiar combination of 26 protons and 30 neutrons happens to be very stable, which means that no more energy can be released either by fission

Fig. 3 Proportion of elements in the Earth.

Fig. 4 Internal structure of the Earth.

S, 135-330 GPa, 3,300-5,800 K, 2,900-5,000 km from the surface), but it is believed that at the core itself, under the conditions of temperature and pressure that prevail there (330-365 GPa, 5,500-5,800 K, d = 10-12), it turns into a solid of a peculiar hexagonal structure, more compact than any of the Face-Centered Cubic and Body-Centered Cubic phases that metallurgists are more familiar with (3) (fig. 4). The outer layer of the planet, however, concentrates other lighter elements, including oxygen and sulphur, which combine easily with iron and other metals. The Earth crust therefore is made of a soup of most of the chemical elements, heavier and lighters ones combined according to the laws of chemistry and thermodynamics and spatially organized by the happenings of geology in their grandest sense. Iron is only the fourth most common element in the Earth crust, at a concentration of 4.6 %, behind oxygen, silicon, aluminum and calcium (fig. 5). Because various processes have been at work over geological times to transport and concentrate some of these elements, metallogenesis has collected metals in specific locations known as ore deposits, which have been later mined by man, if they have been discovered. The most common and useful iron ores are oxides, most noticeably hematite (Fe2O3) and magnetite (Fe3O4) (fig. 6). In the ocean, the concentration of Fe is 2.5 ppb, a very low figure[1]. Before treading into the history of mankind, one must first tell of the cosmic cataclysms where Earth-like planets have exploded and scattered their core in the vacuum of space. These metal meteors, with a quenched structure and a composition in iron and nickel similar to that of the Earth core, are lost forever in space, except for the few that cross a planetary

Fig. 7 An iron meteorite (siderite).

orbit and make it to the surface. These are the siderites (fig. 7), the iron meteorites, which have provided man with his first contact with metallic iron, probably since the dawns of time and, as a material for making various artifacts, in the Neolithic (3500 BC). The role that iron plays as a core material of past and present civilization is directly related to these cosmological and geological storylines, which explain its abundance and, in particular, the large reserves that are still available today as compared to other resources such as oil, for example.

ONCE UPON A TIME IN PREHISTORY

Anthropology defines mankind as the species that invented both tools and abstract metaphysical concepts of the world and of its own place in that world. Tools offered pragmatic solutions for experimenting with and shaping abstract thought, so that both are closely interrelated. Tools are made of materials and materials are extracted from the environment. They are actually abstracted from it in a way that is not completely different from that of shaping philosophical or religious thought. The first materials industry of our Paleolithic ancestors was thus at the same time metaphysics and pragmatic ecology, in their first intrusive interaction with the environment, and economy, offering solutions to meet the needs of the group, i.e. of society. Man the hunter and the gatherer needs tools and weapons to kill his preys, butcher the meat, prepare food, make his cloths and adorn himself. Materials are bones, antlers, teeth, tusks, wood (biomaterials) but also stones, lithic material collected from the ground. Biomaterials are collected from preys, at the same time as meat, but sometimes the priorities may have been turned around : a mammoth might have become more attractive for its tusks than for the organoleptic quality of its flesh.
La Revue de Mtallurgie-CIT Novembre 2004

Fig. 5 Proportion of elements in the Earth's crust.

Fig. 6 Hematite and magnetite iron ores.

940

LE CENTENAIRE

Small groups traveling to find their subsistence had limited needs, which they could accommodate without disrupting their nomadic style of living. The raw materials were thus transformed directly into the final artifact and into the material at the same time. The technology remained simple and based mainly on mechanical crafts, i.e. on cutting, breaking, shaving and polishing. Some tools were used to make the tools of daily use and they used different materials, with different properties chosen to make them perform well in their special tool-for-making-tool capacity. Specialized industries developed as early as 200,000 years BC to prepare flints from rich deposits operated, mutatis mutandis, in a similar way as a modern mine. The technique known as Levallois (from an archeological site in Levallois-Perret, near Paris) of splitting the nuclei into various blades was very efficient in terms of material utilization (high yield) and completely mature within the episteme of the crafts of the time (fig. 8). These are the prehistory equivalent of the modern materials industry, especially in the way that specialized organizations have dedicated themselves to providing materials that would be turned into various tools. The distance between the material and the object that is shaped from it increases with time. Conceptually first, then diachronically as the object is born of the material, and finally in terms of production technology in the Levallois technology the mining of nuclei and the preparation of splinters, that could be given various missions, are carried out upstream and fairly independently of the workshops where knifes or hatchets are made. Biomass and lithic materials could meet the needs of mankind for tens of thousands of years. There was a clear social equilibrium between the technology of the times and the needs of a small population of nomadic scattered groups of huntersgatherers. Exposure to other materials, which were to become more modern, must have happened often, but without creating the driver towards something different during the long Paleolithic era, the Mesolithic and even much of the Neolithic. Technology, though, was brewing with the use of fire to cook food first, to conserve it by drying or roasting and to dry things, change matter and give it new desirable properties. The "art of fire" was used to harden wood, which is probably the first step towards the production of modern materials from earths.

Pottery and ceramics (10th millennia BC in Mureybet), calling for more heat and higher temperatures, as well as the production of lime cement from the roasting of chalk (Serbia, 7th millennia BC) (14) were thus the first ventures of mankind into making complex materials. In addition to breaking and cutting hard and brittle materials, shaving and polishing hard but also softer materials, technology now has added the art of fire to its "process toolbox" and that of shaping things, using soft and malleable materials, which result from mixtures, mainly with water and from some exposure to heat. Both concepts have survived in the slitting process, carried out on steel, hot or cold, with corrugated rolls, but also in the old French word of fenderie, which designated a plant where a bloom or ingot was cut or shaven into pieces or sheets. Wood is commonly shaved in thin layers today, from which plywood is produced. The idea of shaving round steel bars for producing foils has also been contemplated (15), but without much practical success. Metals, which are easy to shape and tough, i.e. do not break easily, came about in steps. Those metals which exist in a native metallic state, because they have a low affinity for oxygen and sulphur, like copper, gold and to a lesser extend silver, of course came first (fig. 9). For this same reason but also because of their rarity in the ground (Cu : 50 ppm, Au : 1 ppb, Ag : 70 ppb), they seem to have appeared historically in that order. The recycling of metals in our modern society is a rediscovery of this early use of metals, as scrap is similar to native metals ; it is also easier to use and leaner in energy needs and CO2 emissions. Technology, not only materials, is thus also recycled! Metallurgy does not simply consist in using native metals, but also in making metals from earths (ores) and in producing new metals by mixing (alloying). Extractive metallurgy (smelting) and physical metallurgy (forging), which are 20th century concepts, were empirically discovered in their craft version during the Neolithic. Native alloys of gold and silver (electrum) were known in Egypt in the 2nd millennia. Copper extractive metallurgy came first, followed by the alloying with other elements: copper was too soft to use as a blade and the addition of small amounts of arsenic and then of tin (5 to 10 % initially, 30 % ideally and eventually) was sufficient to produce bronze, a hard material which sustained for several thousands years as a basic military material (16).

Fig. 8 Levallois technology for preparing flints from lithic nuclei - the first material technology in the history of mankind.

Fig. 9 Early metal artifacts made of copper, bronze and iron.

La Revue de Mtallurgie-CIT

Novembre 2004

This intrusion of metallurgy in civilization accompanies a radical complexification of the economy, with specialized activities in mining, material production, and artifact manufacturing, bound together by trade : The tin needed by the bronze producers of Greece was thus brought from Turkey, Spain, Brittany and Cornwall. This also goes along with the population growth that has started with the settling down of the Neolithic through the discovery of agriculture, urbanization and the organization of larger political areas, which called for tools and weapons on a large scale. One can probably argue that only metals were in position to answers such needs and that metallurgy became therefore a necessity, which led to its invention. The concept of material has by then been fully constructed. It goes along with the specialized work of different trades, carried out at a scale that was more that of various crafts than of an industry but which served the same purposes. The stage was set for more change in terms of materials, as an answer to more needs in terms of volume, properties and cost. Iron was thus the next player to enter the play of history. It was stronger, tougher, potentially more versatile and much more abundant than any other materials used until then. It supplemented and competed with wood and ceramics, a situation that is still true today, when cement, steel and wood are the three most common basic materials produced at the level of about 1 billion tons per year each. Iron slowly displaced copper and bronze, over several hundred years ; in Northern Europe though, copper has remained a construction material for roofs, which can be found as far south as the French city of Metz on the railway station[2]. s

Fig. 10 Steelmaking in Egypt.

tiles provides the heat that increases the temperature and allows the pyrolysis to proceed and transform wood into carbon. A chimney vents the fumes to the outside. The reduction itself takes place in a simple hole dug in the ground (1 m deep and 25 cm in diameter) and eventually in a shaft furnace, which is the ancestor of the modern blast furnace into which it evolved[5], as the horse carriage is the ancestor of the automobile. Ore, charcoal and additives such as lime, clay or sand are piled up in a small furnace, not very different from the charcoal kiln, except for the provision of a chimney, that provides a draft for the air that was injected laterally through a pipe (19, 20) (fig. 10 - 13). Natural ventilation was the norm initially - sometimes an orogenic wind on the slope of a hill, but bellows and water pumps were added in the Middle Ages to supplement natural convection. The air combusted the charcoal into CO, an exothermic reaction that brought the temperature up to maybe 1,100C, a major feat that must have taken hard work in empirically process design to achieve. CO would then take care of reducing the oxide to iron (indirect reduction),

HISTORICAL METALLURGY PREREDUCED IRON AND CEMENTED STEEL

Iron was produced by the reduction[3] of ores (mainly oxides, Fe2O3 hematite or Fe3O4 magnetite, but also (Fe2O3, n H2O) goethite, siderose FeCO3, pyrite FeS2, etc.), found almost anywhere, where man had settled (in lakes and bogs and eventually in the ground (17)). Indeed, until the end of the 19th century, iron was produced in Europe in workshops scattered all over the map. The reducing agent was carbon, artificially produced by converting wood into charcoal. The large forests that had sprung up in Europe in the post-Pleistocene period[4] (18) could thus sustain the needs of man for several millennia until the 18th century. The production of charcoal as well as the reduction of iron involved the building and operation of chemical/metallurgical reactors, where raw materials, carefully prepared, mixed and arranged, were subjected to heat and made to interact. Charcoal kilns, as they are still built in some parts of the world today, have probably not changed much over centuries : wood is piled up and enclosed inside an ad hoc chamber, made of earth and later of bricks, to avoid contact with the atmosphere and provide heat insulation. Some limited air ingress is organized through a small hole where the pile is lit up and the combustion of a small fraction of wood and of the freed vola942

Fig. 11 Schematic diagram of a bloomery, in the stage of development of the early Middle Ages.

La Revue de Mtallurgie-CIT

Novembre 2004

LE CENTENAIRE

Tne (400 BC) cultures. The knowledge traveled also to the East and bloomeries have been known in early China and Korea, and in Japan during the Kofun period (23). Iron and steel were not known in the Americas, which maintained a bronze-based technology until the arrival of the European : the control of iron technology by the Spanish invaders has been one of the factors of their easy conquest of that continent. The bloom was still a rather useless material, which would hardly compete with bronze in terms of hardness. It was necessary to clean it up by removing the gangue and to compact it. This was carried out by forging the hot bloom with a hammer, in the hands of the blacksmith and latter with a forge-hammer moved by hydraulic energy (un martinet, in French). The gangue was beaten out of the bloom by the impacts and the pores were welded shut. In Roman times, a bloom would weigh no more than a few hundred grams. The workshop was eventually called the Finery, Martinet ; Martinette or Moulin (moulines) Fer in France (fig. 14). The bloom was not fully homogeneous and parts of it sometimes exhibited contamination by the carbon of the charcoal, which turned the iron into a harder carbon steel (natural steel). The control of this cementation (i.e. the diffusion of carbon into iron) was carried out in a further process, where the bloom was heated up for long periods of time inside a charcoal bed. This "steeling" of iron, actually a case-hardening, was carried out as soon as Roman times and its first occurrence may be as old as 400 BC (19). But other evidence seems to indicate that sometimes only the sharp edge of a blade would be carbon-hardened, while, in Eastern Europe (Styria),

Fig. 12 Early versions of the bloomery.

Fig. 13 Bloomery and forging.

through the intermediary step of wustite (FeO), while some direct contact of charcoal with the ore performed part of the job (direct reduction). The non-iron compounds of the charge, the "gangue", remained mainly unchanged. The temperature was high enough for the gangue to melt into a slag that could flow out of the furnace near its bottom. The reduced iron, however, would be produced in the solid state and accumulate at the bottom of the furnace as a lump of metal that the English eventually called a bloom[6] ; the furnace itself was a bloomery (21) and a bas-foyer ou bas-fourneau in France (22). Because it accumulated in a furnace region devoid of carbon, it was made of rather pure iron, neither steel nor pig iron, mixed with gangue and full of holes. The process was a batch operation. Of course, no "theoretical" chemistry was available to understand what was actually accomplished in this process : these scientific concepts would not be evolved until the 19th century, after a false start related to the phlogiston theory. Everything was derived by trial and error. The first process to make steel was thus a direct reduction process that produced unalloyed but dirty[7] iron, or rather mild steel. Direct reduction was later all but forgotten when the blast furnace came of age and became the mainstream technology, from the Middle Ages onwards, but it was "rediscovered" in the 20th century when natural gas started to be used as a replacement for carbon. The Hittites of Anatolia in the 2nd millennium BC and possibly before (23rd century ?) have the reputation of having already mastered the art of producing iron : they have given some metal as a present to the pharaohs of the 18th Dynasty (1380-1340 BC). This is much earlier than the Ages of Iron in Western Europe known as the Hallstatt (800 BC) and the La
La Revue de Mtallurgie-CIT Novembre 2004

Fig. 14 The finery.

Fig. 15 Aspect and metallographic structure Damascus steel blades from Japan.

it may have been possible to produce steel (0.4 to 0.9 % C) directly in the bloomery, due to the use of ores rich in manganese, which produced a slag that did not wash the carbon fully out of the bloom (19). The entry of steel on the scene was rapidly followed by the pragmatic mastering of quenching and tempering, which made available a range of hard materials that clearly outclassed bronze. Another breakthrough occurred when the smiths discovered forge welding, thus making it possible to produce larger samples with the sheer strength of their arms. Moreover, it was possible to assemble iron and steel together and thus to produce a kind of pte feuillete (puff pastry dough) by repeating the process over and over again and doing it with twists, both properly and figuratively speaking. This was the art of Damascus steel, mastered as early as the La Tne period, but which retains the name of Damascus, a city in Syria where it was produced for the late Roman Empire. The result is not only a composite material, which is both tough and hard and therefore particularly appreciated in swords, but a delicately crafted piece of steel jewelry, where the pattern of mild and hard steel shows at the surface with various regular geometrical shapes, after polishing and etching. Damascus steels have been produced all over the world, not least in the classical Arab world and in Japan (Katana). The modern equivalent of producing Damascus steel is the production of composite materials but also mechanical alloying. The bloomery, a technology born 2000 years BC in Anatolia, remained conceptually the same until the Middle Ages in Europe. For more than 3000 years therefore, iron was smelted from ore in a bloomery and further "steeled" by cementa-

tion and casehardening. Sophisticated blades were produced by forge welding. Steel technology was thus sustainably based on solid-state transformations. This was slow in terms of kinetics, because diffusion controlled everything. The productivity was small and the cost correspondingly high: Steel was therefore devoted only to strategic applications, the foremost of which was weapon production. This is no longer the mainstream technology for making steel today, but it has not been fully dismissed: Besides prereduction and mechanical alloying, one ought to mention powder metallurgy, which is a contemporary large-scale technology for producing steels in the solid state. Tentative solid state reduction in preformed material has been tempted in the 1980's at the University of Swansea by Professor Singer (24) and the concept also survives for the production of powder by hydrogen prereduction of pure hematite fines (25). On the other hand, the solid state reactions discovered so long ago are still very much alive in the modern technology of thermomechanical processing, which covers everything done on hot metals, and constitutes the core of physical metallurgy, materials science and of much of nanotechnology. s

HISTORICAL METALLURGY PIG IRON AND BLAST FURNACES

The steel technology, that the Roman Empire inherited, was passed on to the new order of the Goths. The bloomery did not change much in most of Europe, busy as it was at looking for the new geopolitical equilibria that would emerge from the exhausted and collapsed Empire. Only in Spain did steel technology strive and progress, under the Visigoths first and then

Stack

The charge Bloom Tuyere

Fig. 16 Early and "improved" Catalan Forge.

944

La Revue de Mtallurgie-CIT

Novembre 2004

LE CENTENAIRE

the furnace had locally become hot enough to reach its liquidus temperature, but too brittle to utilize like the wrought iron that he was used to: The drippings were therefore charged back at the top of the furnace to become honest iron again. It took time and ingenuity to find applications for this pig iron : It started to eat at the niches that the bronze industry had kept, i.e. bells and soon cannons and cannon balls. Thus the industry became ready for the next breakthrough, the Blast Furnace, which appeared initially as a result of incremental innovation. The Blast Furnace was thus a bloomery that produced liquid metal, in a carburized state that distinguished it strongly from the wrought iron and even the steel that has been part of civilization for the last few thousands of years. It was initially called Fluofen in Germany to indicate that the metal was indeed flowing out of it. But it became rapidly the Hochofen and the Haut-Fourneau, while, in England, the bellows were seen as more important, hence the word of Blast Furnace replaced that of High Furnace. The first Blast Furnaces have seemingly been built in the 13th century[8] in Sweden at Lapphyttan (17) and at Jubachstal in Germany, but probably also in Wallonia and in France (fig. 18, 19). An intermediary process capable of switching back and forth between the production of solid iron and liquid

Fig. 17 A bloomery with details of the water-wheel driven bellows (29).

under the Moors. In Catalonia, the antique bloomery built in a hole in the ground and with the heap of raw materials simply covered with clay was slowly transformed into the Catalan Forge, a fitter precursor to the Blast Furnace by its shape (26). The Catalan Forge was a construction built on a hillside, with a hearth stone 30'' square and three stonewalls 3' high (fig. 16). A hole near the bottom let in the nozzle of the leather bellows (27) for introducing the air. The typical output in the 8th century was 350 lbs for a 5-hour heat, to be compared with the 50 lbs that the old technology was only able to yield. The Catalan forge was disseminated all over Europe as the Best Available Technology of the times, but as soon as the economy called for more production, inventors were at work to improve its productivity. The furnace height grew to 8' and then to 16' and the production of each batch jumped to 400 and 600 lbs. This high bloomery was called Stckofen or Wolf Furnace along the Rhine and in Austria, where it originated: The bloom was called a Stck or a Wolf, which gave rise to the French word of Loup. The Stckofen could produce 100 to 150 tons of iron per year surpassing the production capability of a Catalan forge. The increase in the furnace height soon placed the tuyere too high for manual labour to activate the bellows. Hydraulic power, already known since the 12th century (28) was soon put to use to power the bellows as well as the hammer of the blacksmiths (to become the tilt hammer). The idea originated in the silver mines of South Tyrol in the early 13th century and then the ironmaking community got hold of the technology, especially the Cistercian Monks, who spread it with their monasteries all over Christian Europe. The Steel Industry moved from the hillsides and the middle of forests that could provide a steady stream of wood for charcoal, to the riversides of springy streams and rivers (fig. 17). The bellows grew in size faster than the height of the furnace, so that the temperature inside the shaft increased also. Liquid iron started to trickle down the charge to the bottom of the hearth. This did not necessarily please the ironmaker, as the new iron was indeed pig iron, carburized and melted because
La Revue de Mtallurgie-CIT Novembre 2004

Fig. 18 A blast furnace in Europe during the Middle Ages.

Fig. 19 A blast furnace ca the 18th century.

945

pig iron, by changing the fuel rate and the size of the tuyeres, called the Blauofen, coexisted for a while with the old and the new processes. "In the 16th century these furnaces were 6.7 m (22 ft) high and could produce 4,000 lbs of iron per day with a fuel rate of 250 lbs of charcoal per 100 lbs of iron produced. These furnaces had a low life expectancy of approximately 45 days" (26). The displacement of the bloomery process by the blast furnace was the result of the strong economic driver due to the increasing demand for steel and also of the invention of new ways to make iron (wrought) and steel from liquid pig iron, i.e. of the evolution of steelmaking, in the modern acceptance of the word, to accommodate the jump in productivity of the ironmaking processes. This game of pursuit between ironmaking and steelmaking will keep going on until the present and constitute a strong driver for progress. Indeed, it had become simpler to make steel now that pig iron was available in solid and liquid forms. Steelmaking thus evolved over a period extending from the High Middle Ages and the end of the 20th century (19), with a variety of alternative technological solutions : The first concept consisted in heating up layers (sheets) of wrought and cast iron together without contact with air to just below the melting point of cast iron. This is the so-called Chinese process[9] which was invented in China as early as the early 2nd century. Diffusion produced steel, more or less homogeneous, depending on the skill of the operator. One could also introduce bars of wrought iron into liquid cast iron. The bar would absorb carbon by diffusion and then could be forged. The process, developed in Northern Italy, gave rise to Brescian Steel[10]. Heating up layers of wrought iron with powdered charcoal in large sealed chests produced blister steel by the cementation process. It could then be broken into shorter pieces, made into faggots and forge-welded together to make shear steel. Cementation was the major process for making steel in England during the modern period (1700-1850). Blowing air over pig iron melted in a bed of charcoal to remove carbon, a process called "fining" the iron and carried out in the finery furnace (l'affinerie), is the ancestor of the modern steelmaking processes. The product of the finery was a bloom, as complete melting of the steel after its transformation from the cast iron was impossible due to the low temperature (1,400C). It precipitated into small pieces that floated out of the melt and a worker called the Rabbler brought them together with a long steel pole (un ringuard) and much work. The melt was decarburized to less than 0.50 % carbon. A variant appeared in Austria, whereby the process could be stopped at a higher carbon content, thus directly producing a harder, quenchable steel known as Styrian Steel [11]. The process was difficult to handle, labour intensive and full of knowhow[12] as the trick of the technology was to eliminate carbon as CO2 at the top and to avoid excessive recarburization by the carbon hearth. With this technology, the word finery had taken a new meaning as compared to the forging workshop that it used to designate downstream of the bloomery : indeed
946

Fig. 20 Schematics of a puddling furnace.

this former finery has now been split up in two different shops, the new finery, where some liquid metallurgy was carried out and a further shop called the chafery, where the bloom would be heated up and hammered (cingl). The technology of furnaces came to the rescue of the finery in the shape of the reverbatory furnace, a clever devise, where charcoal was gasified in a chamber where the hot flames were reverberated by the roof and transported the heat towards the charge spread out in a second chamber, thus avoiding recarburization to a large extend. The furnace changed into a puddle furnace[13] and the bloom it produced was puddled iron (fig. 20). The material was likely to be heterogeneous in carbon content and in hardness within one "heat" and between heats : low-carbon (fer doux) and high-carbon (fer fort) steels had to be sorted out by hand from the true steel. On the other hand the heterogeneous material was convenient for making blades, as it mimicked the Damascus steel, more painful to make. The final evolution of this technology was the Siemens-Martin furnace, which produced liquid steel directly. The Eiffel Tower is a brilliant example of the achievement of this mid-19th century steel technology (fig. 21). The next breakthrough in steelmaking consisted in obtaining steel in the liquid state, systematically and reliably. This needed a stronger heat source, which was provided by the switch over from charcoal to earth coal - that made available a denser form of energy, and by the improvement in furnace technology. The first man to achieve the breakthrough in Europe was Benjamin Hunstman, a clockmaker who was unhappy with the steel available on the market because of its lack of consistency in composition and properties (30). He had the intuition that if he somehow managed to get the steel in the liquid phase he would obtain a uniform composition that would be carried over to the solidified product. His practical concept was to heat up blister steel in a clay crucible placed in a coke bed properly activated by bellows and he did succeed in achieving melting without direct contact with the hot coke and therefore without recarburization. This was the beginning in 1742 of the crucible process, which was to be used for making special steels well into the 20th century. The induction furnace is actually a modern avatar of this process concept modified to use electricity as a source of heat and ordinary scrap as a substitute for blister steel.
La Revue de Mtallurgie-CIT Novembre 2004

LE CENTENAIRE

The Steel Industry has now discovered some of the major tools for modernizing its operations and for making Steel the flag bearer of the industrial revolution. It will help ushering in the deep transformation of mankind's lifestyle of the 19th ad the 20th centuries, i.e. a massive increase in life-expectancy and standard of living and, as a consequence, lead to the population explosion. Coal was the major instrumental cause of this profound revolution : it allowed the production of larger amounts of cast iron and also that of steel by the crucible process. Process metallurgy is now based on the manipulation of liquid phases at every stage, thus not only deeply changing the old paradigms, but also opening up the way for exponential growth of the production that was to continue well into the 21st century. Until now, we have told a story focusing on European history. Steelmaking technology elsewhere took different routes : in China, the bloomery, which may have come from the Middle East, was rapidly improved in terms of heat efficiency so that the production of pig iron became possible several centuries BC (23). Tradition puts the origin of the charcoal blast furnace in the 3rd century BC, but claims of earlier manufacture have surfaced recently (Zhou Dynasty, ca 550 BC) : plowshares may thus have been cast by and for farmers as early as the 6th century, while the nobility understood its usefulness for warfare only in the 3rd century BC. China knew about coal long before Europe did and the first coke blast furnace appeared there several hundred years before the Darby furnace. The technology was not exported outside of China however and it eventually stopped evolving so that the modern western blast furnace, which had become more advanced, was imported from Europe in the 19th century. In Japan, where iron was only present in sands in very dilute contents (2 % and up to 8 % after beneficiation), a different kind of bloomery called a Tatara ( ), was invented in the 7th century AD (31) and used continuously thereafter, mainly to make swords. The process, shown in figure 22 is a version of the bloomery specially designed to handle low-grade ore : the hearth is wide and low, so that charcoal and sand can be added continuously for the 70 hours of operation, during which the temperature rises high enough to let the slag melt and flow out in large quantities, while the iron is reduced into a variety of products, from pig iron to iron and steel, depending on the physical proximity to carbon. The various irons were separated and recombined to make swords. Only when rich iron ore was discovered in Kamaishi, was it possible to switch back to the western style blast furnace (1857) with a 100 to 1 reduction in operating cost[15]. The evolution of steelmaking technologies in Europe and in Asia, which shows large discrepancies in the timeline of introduction of new processes, is a clear example of how technology development responds to local conditions, both geographic and economic. Change does not follow a linear and univocal path but explores a complex and non-deterministic arborescence.
947

Fig. 21 The Eiffel Tower.

The ingots that Hunstman produced were indeed uniform enough for his needs. He also produced the first castings of steel, thus inventing steel foundry, and sorted out the basic issues of deoxidation. But another breakthrough had happened earlier, when another clever Englishman thought of using coal rather charcoal in the blast furnace and, moreover, to use a pyrolyzed product from coal, coke. In 1708, Abraham Darby leased a small charcoal blast furnace in Coalbrookdale, Shropshire and by 1709 he was producing coke. Over the next ten years, coke was mixed with charcoal in ever increasing proportions until 1718 when iron was produced from 100 % coke as a fuel. Darby did not try to keep the use of the new fuel a secret, but he didnt publicize it either. Up until 1750, the only ironworks using coke on a regular basis were two furnaces at Coalbrookdale and one at Whilley, all operated by the Darby family. Finally, during the period from 1750 to 1771, the use of coke spread with a total of 27 coke furnaces in production. The coke Blast Furnace will replace the charcoal furnace on a steady state basis and at the same time grow in size. The European forest are no longer able to provide a sustainable stream of wood for making pig iron and it was time to find a substitute fuel or to face extinction. This has been done and the risk of extinction has been replaced by the potential of unlimited growth at the historical horizontal of the 18th century at least[14].
La Revue de Mtallurgie-CIT Novembre 2004

Fig. 22 Schematic drawing and old painting of a Tatara furnace.

The number of available technological concepts, however, is finite and the various timelines rearrange this basic alphabet in different combinations springing up at different times. Charcoal and then coal and coke have been providing energy and reducing agents to the steelmakers everywhere on Earth. The pyrolysis of wood or of coal was also eventually used by everyone to deliver a more carbon-like material than was available in nature. The bloomery, low or high, slim or squat, has also been present everywhere in several avatars. s

Siemens (fig. 24). The process was more time-consuming and more in continuity with the concept of the puddled furnace. It also made a kind of steel, which was easier to control in terms of properties as the interaction with the oxygen of the air was less violent than in the converter process. The SiemensMartin Process, which was also known as the Open Hearth Process (OH), rapidly overcame the Bessemer process in production volume and shouldered most of the growth of the Steel Industry until the middle of the 20th century. From then on, the steel industry had been dealt all the cards that were needed to play the steelmaking game in the modern, temporary and mainstream styles that we are familiar with today. One elementary technology was still missing though, because electricity had not yet been invented. As soon as it was available, particularly in the mountains such as the Alps or the Central Mountains in France, an explosion of concepts appeared to use it as an energy source for melting iron, or scrap, as this material was becoming available as a consequence of the growth in steel production. Paul Hroult, who invented the electrolysis cell on which aluminum production is based, also proposed to have an electric arc kindle between electrodes, which led him to invent the Electric Arc Furnace in 1899 (fig. 25).

HISTORICAL METALLURGY - MODERN STEELMAKING, BESSEMER, THOMAS, MARTIN AND HROULT

Coming back to the European timeline, there was still a missing link in the production of steel, as the available production route still required too many steps, including that of the puddling furnace, which would feed material to the chafing plant and then to the crucible plant : metal was not yet liquid all throughout the route, which slowed things down, was very labour intensive and kept costs high. What was direly needed was a true steelmaking plant that would convert cast iron into steel directly, liquid to liquid. Once again, the light came from England, which was advancing fast on the tracks of the Industrial Revolution. Indeed, Sir Henry Bessemer invented the converter, a reactor in which air is blown through the liquid cast iron through tuyeres placed at the bottom of the vessel (32-34) (fig. 23). He was thus the first man to completely achieve a liquid phase production of steel and thus to invent what is commonly known today as steelmaking in a one-step process. It proved capable of converting pig iron (which could soon be used as taped from the blast furnace and henceforth could be called hot metal) in quantities that could eventually range to several hundred tons, to do it fast (the hour was the norm, but modern steelmakers have gone below 30 minutes) and to produce the whole range of very pure steels that is the core of today's steel production. Bessemer's breakthrough was soon followed by that of the Martin brothers in France, who managed to produce liquid steel in a reverberatory furnace modified by Sir William
948

Fig. 23 The Bessemer process as installed in Sir Henry's own steelshop.

La Revue de Mtallurgie-CIT

Novembre 2004

LE CENTENAIRE

Fig. 24 The Open Hearth Process.

Fig. 25 One of the first Electric Arc Furnace built by Hroult and an industrial one in Dommeldange (Luxembourg).

THE STEEL INDUSTRY IN THE SECOND HALF OF THE 20TH CENTURY

We have now travelled to the present, or rather to the postsecond World War Period that gave birth to the present Steel Industry (35).

more recently, environmental issues taken in a broad context, i.e. eco-efficiency and sustainable development, have come to the forefront as the leading driver for change and the Climate Change issue will certainly accelerate this trend (36). Technologies were centered on shorter process routes and continuous processes, but also on technological breakthroughs such as oxygen steelmaking, top, bottom and mixed blowing, continuous casting, continuous rolling such as what the hot strip mill performs, continuous annealing, etc.

Five drivers for growth have driven innovation further in the Steel Industry during the second half of the 20th century (fig. 26) : initially, it was necessary to catch up with the exploding demand due to the need for reconstructing the economies destroyed by war and, in parallel, to cater to the beginning of the consumer era, foremost of which is the birth of a consumer car industry ; once the 30 Glorious Years had gone by, it was necessary to ensure higher and more consistent steel quality ; to increase productivity in order to match the 2.5 % steady yearly decrease in steel prices ; to focus the production more narrowly on customer needs by providing them with tailored-made products, incorporating customer-specific value-added Fig. 26 - Evolution of steel production in the 20th century showing technology that are usually called "steel solutions" or the drivers behind the deep changes that took place after the 2nd World War. "solutions in steel"[16] ;
La Revue de Mtallurgie-CIT Novembre 2004

949

In parallel, prereduction, based on the use of natural gas grew to occupy a niche of roughly 10 % of the iron ore processing routes[17] and, because of the accumulation of steel and its leakage back into the economy as scrap, the EAF has grasped a share of steel production as high 43 % (37). The OH process waned slowly only, as it still accounted for 3.6 % of world steel production in 2003, with 30.1 % of the total production of the former USSR countries and some more furnaces left in India (5.7 %). On the other side of historical development, new processes have appeared, while even more have been studied but did not yet pass the glass boundary that separates R&D from commercial implementation : smelting reduction and near-net-shape casting have been the major families of endeavour at this point in history. Figure 27, conceptually borrowed from (38), shows the historical evolution of the steelmaking processes in two coupled Fe-O and Fe-C binary diagrams. Native metals and solidstate reduction have provided a direct route to making steel (bloomery) in the early days of history and well into the European Middle Ages. The use of carbon, from charcoal and from earth coal, has made it possible to explore the liquid metal paradigm and its strong advantages in terms of high output and high purity : initially, liquid steel was out of reach, but the mastery of heat transfer made it possible to produce it, with a variety of concepts that culminated in the converter and especially the oxygen converter (BF & Blister steel, BF & Crucible or Bessemer Steel, BF and modern BOF). Natural gas has revived the prereduction direct steel concept and the availability of electricity has made it possible to obtain liquid steel without going through the pig iron stage. This is the point in history where we stand now. s
Fig. 27 Historical evolution of steelmaking processes.

the future growth of capacity might be based on this mainstream technology with a strong emphasis on products and steel solutions rather than on new steelmaking technologies. This certainly looks like what is happening in China today, but is by no means certain in the longer term. Indeed, the sustainability driver is becoming stronger than ever before, especially in its environmental component, and the Global Warming issue is probably the most pressing one. It will remain pressing for a long time. In such a context, the Steel Industry would need technologies that have the potential of reducing GHG emissions by the "factor 4" that European governments are setting as targets for 2050 and this will make it necessary to re-examine this efficient mainstream steelmaking technologies of today and refocus them on lowering GHG gas emissions while maintaining the same energy efficiency level. There are currently three ways of addressing this question. One consists in staying close to the carbon-based Blast Furnace technology and to ensure that the ensuing CO2 is captured and sequestrated; this would lead to deep changes in process in order to make CO2 capture easier to achieve, such as recycling the BF top gas after decarbonatation and thus operating the furnace under nitrogen-free conditions. Another solution would consist in replacing carbon by other reducing agents and fuels, such as natural gas, hydrogen or electricity. This might open the way to using electrolysis for producing steel, a proposal that was never seriously examined in the past because of the high cost of electricity. On the other hand, electricity or hydrogen would have to be produced by dedicated generation plants based on CO2-free fuels, i.e. renewables, fossil fuels with CO2 capture and sequestration or nuclear. A third solution would be based on the use of biomass, which contains "short-cycle" carbon and would not contribute to GHG emissions because a steady-state production of biomass by agriculture or forestry would be included in the carbon loop.
La Revue de Mtallurgie-CIT Novembre 2004

THE STEEL INDUSTRY TOMORROW

The 21st century has started with a strong acceleration of Steel production due to the steady demand of the fastest growing Asian economies, China and India, but also of South America and Eastern Europe. These enormous increases in capacity are obtained by introducing the newest and best technology available on shelf from plant builders but also by improving the productivity of existing facilities beyond the state-of-the-art as defined in Japan or Europe. On the business side of things, mergers have brought to the market players with a capacity in excess of 40 Mt/annum, such as Arcelor and LMN. Moreover, the industry seems to have switched over from a supply to a demand-driven sector and therefore the price of steel has stopped declining, at least in the short term. Does this foretell of incoming change of steelmaking technologies or, on the contrary, will this mean the end of change in this area ? This is by no means a rhetorical question, as the technology legated by the 20th century has reached a scale and a proximity to thermodynamic limits that makes it difficult to continue changing at the same pace as it did in the past centuries. For example, 10 to 15 % improvement in energy needs per ton of steel is the most that can be forecast. The Steel Industry has matured and reached an asymptotic level of technology and
950

LE CENTENAIRE

Various programs are being started around the world to address these issues and come up with a commercial solution within the next 10 years. One of them is the Europe-wide ULCOS[18] project (39) but others are under way elsewhere coordinated from IISI in a "CO2 breakthrough program" (40). Any solution that comes out of this program would be a deep paradigm shift in steelmaking technologies as compared to the present routes. The new ones, which would at the beginning probably be more expensive to operate than the mainstream technology, would compensate this extra-cost by added environmental value, translatable into cash through CO2 trading markets or CO2 taxes. Whatever direction is eventually chosen, it will prove a major breakthrough, on a par with those that took place in the long historical past that we have reviewed here. In the longer term, the world will face other societal and geopolitical changes. Eventually population growth will stop and, when this happens, the reliance on virgin iron will not be as strong as it was in the past: Recycling will be in a position to provide all the more of the iron units as the recycling rate will be high. As much as 70 or 80 % of the iron units could thus be provided by scrap. The economy will also dematerialize, because materials will be re-used, recycled and used in consumer goods with longer lives and because the intensity of artifact use in the economy may also decrease. Design for Recycling (DfR) and Design for Durability (DfD) will pull in this direction. This will also help decrease GHG emissions[19]. Another important change is that the technology trend, which has been pointing towards larger output, more complexity and upscaling, will branch off to another direction : simpler, more frugal and rustic solutions will be preferred in parts of the world which are presently emerging into development. "Less is beautiful" will become a new slogan. In terms of steelmaking technologies this may mean that smaller plants, in equilibrium with local conditions, will appear as a countermodel to the large mainstream integrated mill of today, a path that the mini-mills have been treading since the 1960's. There is for example no reason to believe that the 5 Mt/annum mill that is the baseline "best" solution today would remain the best solution for an electrolysis-based steel mill. Other concepts based on the use of local raw materials are likely to be developed, provided that the new technology remains rustic and cheap to develop: A small blast furnace, either fed with charcoal or non-coking coal, in combination with a simple

prereduction process (e.g. rotary kiln, carbon-based) could become very successful in countries like India. Such a deconstruction of the steelmill model will probably be matched by a deconstruction of the steel market. The concept of "steel solutions" is the first step in this direction and has proved a major driver for generating economic value and for steering the creativity of the steel industry downstream of its present core-business. Among the longer-term trends, we might expect a departure from the comodification of steel that would be driven by aesthetic values under the influence of architects, artists and designers. Modern automobiles are beautiful because of the rounded shapes that deep-drawing steels allow. Weathering steels change aspect and surface texture with their ability to accommodate oxidation, a smart material behaviour. s

THE STEEL INDUSTRY IN THE FAR FUTURE

The implicit statement that we have taken for granted in the previous section is that steel will continue to be needed and used by mankind long into any foreseeable future. Indeed, no material yet in history has proved as sustainable as iron and steel and there is no sign of any emerging material that can compete with it at a global scale. We have agreed that this stems from the abundance of iron as an element, which is due to the cosmological production of iron in the star furnaces. This is likely to be very enduring. A consequence is that a shortage of iron will probably never happen unlike what is probable for fossil energy. The iron at the core of the earth is probably forever out of reach. Maybe mankind should plan to recover it when the Sun expands into a red giant rather than let it seed the Galaxy for the benefit of future planets and civilizations. This however would constitute very-long term planning! Meteoritic iron on the other hand would probably be within reach of our present technology, if we so chose to allocate our resources. It is interesting to imagine what would be a spacebased technology (41) : it could use nickel-bearing metallic iron as iron sources, the Sun as the energy source, surface tension as the container for liquid steel and 3-D Urchin Continuous Casting to directly cast net-shape long and flat products at the same time. Using the material in orbit would probably be more cost effective, but steel could also be floated down planetside on giant gliders or ferried down in the space elevator (42).

Fig. 28 A panoramic view of Mars from the Sagan Memorial Station where Pathfinder landed in 1997 (the red rock is mostly hematite) (43).

La Revue de Mtallurgie-CIT

Novembre 2004

951

In interplanetary space, iron would be the material of choice, because of its availability. On other planets, if mankind ever settles on them, this might still be the case. For example, iron ore is everywhere on Mars (fig. 28). NASA is considering various systems for producing hydrogen and the electrolysis of iron oxide is one of the explored routes, which would also produce iron as a by-product (44). A more immediate issue is energy shortage. The peakoil is predicted soon, 15 to 30 years from now. This will drive the price of energy to such heights that energy demand will peak in the 2040's or slightly earlier, depending on whether there is a carbon constraint (45) and assuming that renewable energies are developed at the a brisk pace[20] that coal is used more than it is today and that any energy still needed is provided by nuclear power[21]. Cosmologists like to talk about the arrow of time to designate an irreversible and deterministic course of events ; there is however no such irreversibility in historical trends and in exercises in foresight or technological forecasting. Short of determinism, some invariants are however likely to be observed: The needs of the economy and of society will continue to drive the offer and the choice of technologies. Moreover, new technologies will be born from the accumulated wealth of know-how and knowledge accumulated over historical times, a fact that is recognized when it is said that the Steel Industry is based on Cumulative Technologies. One strong invariant is that steelmaking will continue to be based on processes involving a liquid steel phase, everything else remaining subject to change.

CONCLUSIONS

Iron is a very common element in the universe and on planet Earth, which is mainly made of metallic iron. This fact is due to the particularly stable nature of the iron nucleus that directs stellar nucleosynthesis towards producing iron as the final byproduct of nuclear fission. The reserves of iron in the Universe are enormous and growing, as the stellar furnaces are still operative. They amount to 1045 t (1018 Suns). The Earth itself contains about 1020 t of iron or enough raw material to ensure the present production of steel (109 t/annum) for as long as 7 times the age of the Universe! In the crust, the amount is estimated at 1017 t, while the world mining reserves are estimated at 230 billions of tons (46). Cosmology, geology and mining projections agree on the fact that iron will remain widely available for any foreseeable future Not only does the nucleus of the iron atom exhibit special properties that explain the previous conclusions, but its electronic cloud also has special features which have turned it into a necessary trace element in living organisms and into a material that exhibits a rare combination of strength, ductility and toughness and an ability to combine with other elements in alloys that constitute the largest family of inorganic compounds known to Chemistry, Materials Science and Metallurgy. This combination of ready availability and of outstanding properties explains the enduring presence of iron and steel in the history of mankind and guarantees that it will continue to accompany it in the future for a very long time indeed. Man has learned to domesticate iron in ways related to mining and metallurgy. The development of steelmaking technology

Fig. 29 A timeline of materials and steelmaking technologies over the last 200,000 years.

952

La Revue de Mtallurgie-CIT

Novembre 2004

LE CENTENAIRE

Fig. 30 Production routes for steel in 2050. In light-blue rectangles are shown technologies, which will need to be fully developed until then.

has been driven by the demand for volume production, which has followed the increase in population and the material intensification brought by "civilization". Processes have been scaled up at the pace required by history and the scale-up has been accompanied by a reduction of costs : the material became cheaper as it became more widely available. Another strong enabler of this rise in volume has been the switch from solid to liquid metallurgy with the use of high temperatures. This in turn allowed access to better-defined materials, with consistent compositions and properties. Besides this hard "economic" explanation, a softer socio-historical approach should emphasize the point that the needs of society are also a search for more refined concepts for describing the physical world and that this intellectual need cannot be distinguished from the economical need (47). If we look closer at the technology timeline (fig. 29), we find a tree-like structure : change or progress follows one branch until a particular technology exhausts its promises and the branch stops. Breakthroughs are like new twigs, branching off in different directions originating from new concepts. We have currently been exploring a solid, strong and long branch of steelmaking technology, but it looks like we are nearing its limits : mainstream technology has reached a kind of frozen perfection. The driving force for this momentous change and also for this evolutionary cul-de-sac has been the immersion of the Steel Industry in the free market consumer economics of the end of the 20th century and the globalization it brought along. Is this the end of the story or the prelude to the birth of a new technology branch? If the past gives any clue to the future, picking up the first answer would be foolhardy : the end of history (48, 49) has been predicted many times without much empirical confirmation The steelmaking branch that the present Steel Industry has been built on will remain strong and might even grow thicker, but it will probably not grow longer. On the other hand, new branches are likely to sprout, driven by two outside drivers, which will act as paradigm shifts :
La Revue de Mtallurgie-CIT Novembre 2004

global environmental issues such as Climate Change will probably pull technology and steel-making in new directions; and the end of population growth, expected some times towards the end of this century, will channel human creativity towards post-modern societies, where the creation of value will not be based on material production nor on service, but on subtler deliverables related to pleasure and aesthetics. The toolbox of robust technologies (41) that is available as the result of the cumulative nature of steelmaking will probably be put to use ever more eagerly, in the same way that modern physics has learned to weave robust enduring concepts like quantum mechanics and general relativity into new yarns like Superstring or M theories. The keys to the future can therefore be read in the past with appropriate insight. A somewhat more explicit projection of what the future steel production may look like is shown in figure 30 showing the likely steelmaking processes in 2050 (41, 50). The invariants, as compared to today's technologies, are the following : Iron units stem from ore and scrap. Liquid steel is processed either in an Oxygen Converter or an Electric Arc Furnace. Solidification is carried out continuously on a series of continuous casting processes that produce a range of thickness from slab to thin strips. Coated and composite materials are produced downstream in specialized plants. Changes are likely to occur in the following areas : The reduction of iron ore will be based on a mixture of reducing agents, from coal to electrons, with natural gas and hydrogen as new players. The Blast Furnace will change to accommodate capture and sequestration of CO2. The iron ore/scrap ratio will be tipped to the scrap side.
953

(5) (6)

www.rcs.org , Protein data bank. COALE (K.-H.) et al Southern Ocean Iron Enrichment Experimen t: Carbon Cycling in High- and Low-S, Science 304: 408-414 (2004). BISHOP (K.-B.) et al Robotic Observations of Enhanced Carbon Biomass and Export at 55 degrees S D, Science 304: 417-420 (2004). BUESSELER (K.-O.) et al. The Effects of Iron Fertilization on Carbon Sequestration in the Southern Oc..., Science 304: 414-417 (2004). BLAIN (S.) La Recherche, n 376, 12-13 (June 2004). SEGUIN (M.), VILLENEUVE (B.) Astronomie, Astrophysique, Masson (1995). REEVES (H.) Dernires nouvelles du Cosmos, vers la premire seconde, Seuil (1994). GREENE (B.) The fabric of the cosmos, Alfred A. Knof (2004). TROMPETTE (R.) La Terre, Belin Pour la Science (2003). MOHEN (J.-P.), TABORIN (Y.) Les socits de la Prhistoire, Hachette Suprieur (1998). Shorter routes to thin sheets and strip, the Metals Society (1982). MARGURON (J.-C.), PFIRSCH (L.) Le Proche-Orient et lgypte antiques, Hachette Suprieur (2001), p. 95-98. Iron & Steel, today, yesterday and tomorrow, a conference to celebrate the 250th anniversary of Jernkontoret, vol. 3 (June 11-14, 1997). RILEY (C.-L.) The origin of civilization, Southern Illinois University Press (1969). BARRACLOUGH (K.-C.) Blister Steel, the birth of an industry, The Metals Society (1984). BURTEAUX (M.), CORBION (J.) Le savoir fer, Glossaire du Haut-Fourneau, Association Le Savoir Fer, 57Florange, Presses du Tilleul, 4e dition (2003). SACHSE (M.) Damascus Steel, Stahl und Eisen (1994). http://www.sollacmediterranee.com/decouverte/histoire/histoire.htm WAGNER (D.-B.) The earliest use of iron in China, in Metals in antiquity, ed. by Suzanne M. M. Young, A.M. Pollard, P. Budd and R.-A. Ixer (BAR international series,792), Oxford : Archaeopress (1999), p. 1-9. AHNIER (S.-N.), SINGER (A.-R.-E.) Ironmaking and Steelmaking n 3 (1983), p. 135-141. wwwepma.com/rv@pm/PM%20Process WAKELIN (D.-H.), RICKETTS (J.-A.) The Nature of Ironmaking, in The Making, Shaping and Treating of Steel, 11th Edition, Iron-making Volume. FISHER (D.-A.) The Epic of Steel, Haper & Row, NY (1963). http://lamop.univ-paris1.fr/W3/lamop1.html http://moulinafer.free.fr/[email protected] BARRACLOUGH (K.-C.) Crucible Steel, the birth of a technology, The Metals Society (1984).
La Revue de Mtallurgie-CIT Novembre 2004

(7)

(8)

(9) (10) (11)

(12) (13) (14) (15) (16) (17)

New and revived concepts like direct steelmaking, electrolysis and continuous steelmaking will lead to commercially important technologies. Near-net-shape casting ought to catch a larger market share, when older production capacity needs to be replaced, and it will move towards net-shape casting (thin strip casting, wire casting). There is another context, though, where the liquid phase metallurgy paradigm could be broken. It is related to transmaterialization and to nanotechnology (fig. 31). Indeed, liquid phase metallurgy has the objective of easily achieving the spatial mixture of atoms that equilibrium allows. Nanotech on the other hand proposes to directly arrange atoms at the atomic scale, one at a time. Today, this is understood as a technology in the making that opens up new frontiers at very small scales to make tiny objects that could be small enough to apply the mechanical paradigm to the realm of life sciences. More forward looking thinkers dreamers ? talk about self-replicating machines, which would take care of the difficulty of mass producing these machines. The production of bulk materials by nanotechnology seems mainly out of reach today, because of time limitations (kinetics), energy need and cost unless these self-replicating systems bring about a solution. One might even say that classical technologies are nothing less than large-scale nanotech : the Steel Industry thus operates at the scale of 1038 atoms per year s
(1) (2) (3) (4)

(18) (19) (20)

(21) (22) (23)

(24) (25)

REFERENCES
ARISTOTLE Seconds analytiques, II,11. EMSLEY (J.) Nature's building blocks, an A-Z guide to the elements, Oxford University Press (2001). FOX (B.-G.) Cofactors and Coenzymes, www.biochem.wisc.edu/biochem625/index.html FOX (B.-G.) Iron cofactors: non-haem, Encyclopedia of Life Sciences, A668, Macmillan Reference Ltd (2001).

(26)

(27) (28) (29) (30)

954

LE CENTENAIRE

(31)

TATSUO INOUE - Science of Tatara and Japanese Sword, Traditional Technology viewed from Modern Science, ICBTT 2002, The 1st International Conference on Business & Technology Transfer, Kyoto, Japan (October 20-21, 2002). BIRAT (J.-P.) The relevance of Sir Bessemers ideas to the Steel Industry in the 21st century, First Bessemer Lecture, London, Ironmaking & Steelmaking, 2004, vol. 31, No. 3, (September 2003), p. 1-7.. BESSEMER (F.-F.) The Technology and the Times, Steel World, vol. 3, No. 1, 11-18. BESSEMER (H.) Sir Henry Bessemer, FRS, an autobiography, London, The Institute of Metals, reprint (1989). BIRAT (J.-P.) Sustainable steelmakingparadigms for growth and development in the early 21st Century, La Revue de Mtallurgie-CIT, 98, n 1 (2001), p. 19-40. The European Commission, Steel Technological Platform, Brussels (12 March 2004), ftp://ftp.cordis.lu/pub/coal-steelrtd/docs/steel_stp@[email protected] http://www.worldsteel.org/wsif.php EKETORP (S.) Can we find radically new ways to make iron and steel, the 6th Iron and Steel Congress, Nagoya I (21-26 Oct. 1990), p. 335-341. BIRAT (J.-P.), HANROT (F.), DANLOY (G.) CO2 mitigation technologies in the steel industry, a benchmarking study based on process calculations, Scanmet II, 2nd International Conference on Process Development in Iron & Steelmaking, Lule (6-9 June 2004), p. 73-80. www.americanrecycler.com/nov03/world.html BIRAT (J.-P.) Innovation Paradigms for the Steel Industry of the 21st Century : Future Directions for Steel Industry and CC, Manfred Wolf Symposium, Zrich (2002). www.isr.us/spaceelevatorconference/ http://science.nasa.gov/headlines/y2001/[email protected] KHETPAL (D.), SADOWAY (D.) From oxygen generation to metals production: in situ resource utilization by molten oxide electrolysis, NASA Microgravity Materials Science Conference, Huntsville, Alabama (June 25-26 2002), //msad.msfc.nasa.gov/matsc2002/ CRIQUI (P.) EPII-CNRS, Grenoble, personal communication (2004). http://minerals.usgs.gov/minerals/pubs/commodity/iron@ ore/340302.pdf LATOUR (B.) Pasteur : guerre et paix des microbes, Paris, La Dcouverte (2001). FUKUYAMA (F.) The end of History and the last man, Penguin Books (1992). ASIMOV (I.) The end of eternity (1955). MANNING (C.-P.), FRUEHAN (R.-J.) Emerging Technologies for Iron and Steelmaking, JOM, 53 (10) (2001), p. 20.

[3]

a "chemical" word, that states that the weight of the ore is reduced in the process of turning it into a metal. and (anecdotically?) terminated the large herds of wooly mammoths and European rhinoceros. a 12th century blast furnace in Sweden produced 100 kg of pig iron per day, as compared to 10,000 t/day in a large modern one today, a 105 increase over 700 years. in French, loupe, massot, masseau, saumon, mazelle, renard and loup; the word bloom has a different etymological origin from the homonym that means flower : the bloom, therefore, is not the flower of the furnace. in the sense of non-metallic inclusion cleanliness, although the level of contamination was several orders of magnitudes larger than for a modern steel. the first blast furnace built in North America was at Falling Creek, Virginia in 1622. also called co-fusion, although no melting was taking place. 1% C and 0.8 % oxides. or German Steel, or Cullen Steel (for Cologne, the City through which it was exported), or even, confusingly, natural steel. the Styrian, for example, used a manganese rich iron ore, which produce pig iron with 5-10 % manganese, known as spiegeleisen (because it showed when broken flat grains similar to mirrors). The spiegeleisen was easier to reduce because of the intermediary manganese oxides that formed early in the process. invented by Henry Cort (1784). direct quote from reference (19). the Tatara process, which is closely related to the technology of making swords in the traditional Japanese way, a technology preserved as a "living" museum, has been recently revived and a furnace has been build by the Nippon Bijutsu Token Hozon Kyokai, in Torigami, Shimane Prefecture, in cooperation with Hitachi Metals in 1977. It provides 3-4 tons of steel every year (3). Laser-welded blanks, hydroforming, steel construction, maintenancefree steels for shipbuilding, etc. what distinguishes the modern prereduction processes from the old bloomery process is the use of liquid steel metallurgy to separate steel and gangue : there is no need for brute mechanical force anymore to maneuver hammers or tilt hammers. Moreover, natural gas is used as the preferred reducing agent in a shaft continuous furnace ; which does evoque the bloomery, which is its historical ancestor. Coal-based prereduction has not disappeared, but is carried out in rotary-kilns. Ultra-Low CO2 Steelmaking. Ehrlich & Holdren's equation : , where CO2 is the anthropogenic CO2, Tsteel is the steel production, GDP is the Gross Domestic Product and POP is the population size. including, possibly, solar towers. on the other hand, it does not take into account longer-term solutions, such as Nuclear Fission or Space-Based Solar Energy, beamed back by microwave to the Earth surface. The projections are for them to become available beyond the middle of this century at the earliest, which is a very far forecasting horizon. Such energy would probably be limitless but also quite expensive in terms of capital expenditure.
955

[4]

(32)

[5]

(33) (34) (35)

[6]

[7]

[8]

(36)

[9] [10] [11]

(37) (38)

[12]

(39)

[13] [14] [15]

(40) (41)

(42) (43) (44)

[16]

[17]

(45) (46) (47) (48) (49) (50)

[18] [19]

[20] [21]

s
[1] [2]

AUTHORS NOTES
Na is for example at the concentration of 1.05 %. and Louis XIV had copper plates bought from Sweden to cover the roofs of the Palace of Versailles.
Novembre 2004

La Revue de Mtallurgie-CIT