Coal Utilization in the Steel Industry

By Gareth D. Mitchell, Director
Coal and Organic Petrology Laboratories
The Pennsylvania State University


The largest single use of coal in the steel industry is as a fuel for the blast furnace, either for the production of metallurgical coke or for injection with the hot blast. Other less commonly thought of uses of coal is for making steam and electricity, as sources of carbon addition in steel making processes, and in direct smelting of iron processes. Furthermore, electricity purchased from outside sources is largely generated from pulverized coal combustion and therefore has an indirect influence on steel making operations.

Except for coke making, the requirements for a quality coal product are fairly simple. For pulverized coal combustion, whether taking place in a combustion unit or in the blast furnace, the coal must deliver a known and consistent calorific value, be reasonably low in ash yield or have a relatively benign ash chemistry and meet environmental standards for sulfur and nitrogen oxide emissions. In addition, it must be relatively easy to grind and to handle.

The requirements of coals purchased for coke making are much different from those used in other processes. Only a certain class of coals possessing very specific properties and composition are suitable for the making of a quality coke for blast furnace use. In what follows, these differences will be discussed from the point of view of coal formation, characterization and classification in context with their use in the various steel making operations.


Coal is a readily combustible rock containing more than 50% by weight or more than 70% by volume of carbonaceous material including inherent moisture. It is formed from the compaction and induration of various altered plant remains similar to those found in peat.

Peat consists of plant debris that, for the most part, was accumulated in place in swamps, marshes or bog environments from the different parts of living plants, i.e., their roots, stems, leaf materials and reproductive parts. At a more basic level, plants are composed of various membranes and substances that are chemically distinct. Some of the more important ones are carbohydrates (starch, cellulose, lignin), protoplasm, chlorophyll, oils, seed coats, pigments, cuticles, spores and pollen, waxes and resins. In addition, degradation products, bacteria, fungi, invertebrate and vertebrate organism also may contribute to the organic fraction of a peat. Because of the depositional environment, inorganic components can be deposited with the accumulating organic debris or may form in place by chemical reactions or be fixed by bacteria or the plants themselves. Consequently, peat is composed of water, carbon, hydrogen, oxygen, nitrogen, sulfur, mineral matter and trace elements and contains the building blocks from which coal is formed.

For the most part, coals that are utilized today were deposited as peat many hundreds of millions of years ago from plants that are now extinct or that represent an insignificant part of today's flora. However, the processes of accumulation and preservation of such a wide variety of organic materials have changed little over geologic time. The environmental setting requires that there be high organic productivity and slow continuous subsidence that insures that the peat is protected from inundation by inorganic sediments or from becoming dry and exposed to oxidation. Studies of modern peat-forming environments show a succession of plant communities exist in any given place that varies with time vertically and laterally depending upon local conditions. These different communities, including aquatic plants, reeds, sedges, forests and mosses, contribute variable amounts and types of organic matter to the peat. The processes that influence changes in plant communities and the plants themselves account for the heterogeneous chemical and physical properties of the resulting coal.

Following deposition and in the early stages of burial, plant remains undergo biochemical coalification principally from the action of bacteria and fungi. This process of humification largely alters the chemical nature of the organic materials and may continue for sometime after burial, but ultimately gives way to factors involved in geochemical coalification. This stage of coal development is controlled primarily by the rise of temperature in response to the geothermal gradient as organic material becomes buried at greater depths, the time over which this process occurs and the pressures to which it is subjected. Thus, the types of plants and plant tissues contributing to the original organic deposit and the action of biological and geologic events during and after burial are responsible for the great diversity of the resource we refer to as coal.


Direct and indirect utilization of coals for production of energy and chemicals as well as for smelting of base metals is the foundation upon which our interest in classifying this resource is built. However, because of their complex, heterogeneous nature and the variety of coals used throughout the world, classification is a difficult task that is both time-consuming and expensive. Identification of the most advantageous raw material, whether by quality, cost, availability or some combination of these factors has always been one of the driving forces behind the development of classification systems. In fact, many of the systems in use today were derived specifically from a need to identify quality coals for coke making, and in that respect only classify a relatively narrow range of coals. Other systems that have been developed to address the scientific need to understand the origin, constitution and fundamental properties follow the approach that any sound classification will identify all coals for all potential industrial uses.

Generally, coals are grouped according to particular properties as defined by their "rank" (degree of metamorphism), "type" (constituent plant materials) and "grade" (degree of impurities and calorific value). Of these, rank is a fundamental concept that involves a qualitative expression of the coalification sequence and is universal to all classification schemes. Coalification is a term that describes the maturation of plant tissues from peat through different stages of lignite/brown coal, subbituminous and bituminous coals to anthracites and meta-anthracites. As demonstrated in Figure 1, many chemical and physical properties change during this progression, but unfortunately there is no single property that changes uniformly over the complete range. Furthermore, a coal's type and grade influence many of the measured rank parameters.

The types of analytical procedures needed to characterize and classify coals can roughly be divided into those that describe chemical composition/properties, petrographic composition and those that describe mechanical/physical properties. Some of these procedures are basic to the evaluation of all coal materials, whereas others are employed in the evaluation of their use in specific processes, like coke making.

Classification of coals requires some methodology for measuring their chemical, physical and industrial properties. For this purpose, a variety of standard analytical procedures are available world wide from reputable standards organizations. Some of the more familiar ones include the International Organization for Standardization (ISO), the American Society for Testing and Materials (ASTM), the British Standards Institution (BS), the Standards Association of Australia (AS), the Association Francaise de Normalisation (AFNOR), etc. Because these organizations may support different standards for the same test, the standard practice used always should be identified, but more importantly the procedures described in these standards should be followed to the letter. For the purposes of this discussion, those methods maintained by ASTM will be adhered to [see: American Society for Testing and Materials (ASTM), 1999, Annual Book of ASTM Standards; Section 5, Gaseous Fuels; Coal and Coke, vol. 05.05, American Society for Testing and Materials, Philadelphia, Pennsylvania, 494 pp; Web Site:]

a) Rank

The classification system used in North America and that is fairly universal is maintained by the American Society for Testing and Materials (ASTM) and designated D388. This approach uses standard methods of measuring and reporting the calorific value on a moist, mineral-matter-free basis and volatile matter and fixed carbon on a dry, mineral-matter-free basis to classify coal rank. Implicit in this characterization is the fact that the amount of moisture, yield of ash and total sulfur must be measured to obtain the correct basis of comparison. Table 1 defines the rank classes and groups with regard to the appropriate ranges of calorific value, volatile matter and fixed carbon, whereas Figure 1 demonstrates how these values change with increasing metamorphism and compares them to other rank measuring parameters, such as moisture and maximum vitrinite reflectance. Figure 1 also helps to explain why the ASTM method employs different parameters to classify coals. For example, calorific value varies linearly with increasing rank from subbituminous to medium volatile bituminous rank and then becomes unreliable. Volatile matter content on the other hand is quite variable until medium volatile rank and then changes sufficiently to become a good measure of the degree of coalification through anthracite coals.

Another very important rank parameter that is useful in the characterization of coking coals is the measurement of the mean maximum vitrinite reflectance. As shown in Figure 1, this analytical technique is sensitive, persistently changing throughout coalification and is particularly important for accurately measuring minor differences in those coals used for coke making, i.e., high volatile A bituminous through low volatile bituminous. The technique measures the amount of incident light reflected from a polished surface of the main component of coal, vitrinite. Vitrinite is that component of coal principally derived from woody tissues and, at least in coals from North America, represents the dominant component.

Rank is the most fundamental concept relating both coalification history and utilization potential of a coal. Volatile matter and maximum vitrinite reflectance are important values used to determine the worth of coking coals. However, because volatile matter is dependent on both rank and composition, coals of different composition may be assigned to the same rank value even though their levels of maturity may differ.

b) Type

As discussed previously, coal is composed of the sum of all the organic vegetable matter preserved and buried as peat. Changes in the chemical and physical properties of whole coal are the summation of changes to the coal constituents. There are three main groups of materials that constitute coals and that are used to define coal type. These material groups identified under an optical microscope in reflected white light are referred to as vitrinite, liptinite and inertinite and are composed of individual constituents called macerals. The three maceral groups are characterized by materials that belong together because of their similar origin or mode of preservation as well as by their gross chemical composition.

In general, vitrinite group macerals are derived from the humification of woody tissues and can possess remnant cell structures or may appear structureless. Typically, this material contains relatively more oxygen than the other macerals at any given rank level. The vitrinite group macerals are usually the most abundant maceral group occurring in higher rank coals. Liptinite group macerals are derived from plant resins, spores, cuticles and algal remains that are fairly resistant to bacterial and fungal decay. They are characterized as having higher hydrogen content than the other macerals, particularly at lower rank. However, at the boundary between subbituminous and bituminous coal there is a marked decrease in their volatile content and increase in carbon. By medium volatile rank a further decrease in hydrogen and volatile content occurs which makes them nearly indistinguishable from vitrinite. The inertinite group macerals are derived mostly from woody tissues, plant degradation products or fungal remains, and are characterized by a high carbon content resulting from thermal or biological oxidation. Inertinite group macerals are found in variable abundance in coals, but are characteristically higher in those from the Southern Hemisphere.

With regard to coke making, vitrinite macerals constitute the principal reactive components of a coking coal. That is, during heating in a reducing atmosphere, vitrinite will become plastic, devolatilize and then solidify to form the porous, carbonaceous matrix of a metallurgical coke. Liptinite macerals are also highly reactive during coke making, but owing to their higher volatile content they contribute more to the by-products than to the coke product. Inertinite macerals are basically inert during the carbonization process, as they do not possess or have only limited thermoplastic properties and volatile contents. However, they do serve a very important function as a filler phase for the other reactive macerals of coal. Small-size inertinite particles thicken the walls between vacuoles in coke thus improving the overall strength of the coke. Consequently, an understanding of the composition or type of coal can be very important to evaluating the quality and value of a coking coal.

c) Grade

Coal grade is a term used to indicate the value of coal material as determined by the amount and nature of ash yield and the sulfur content following the complete oxidation of the organic fraction. Calorific value is one of the principal measures of a coal's value as a fuel and is directly influenced by mineral impurities. Coal mineralogy is not only important to combustion characteristics, but also as materials that can be passed on to secondary products such as metallurgical coke. Alkalis-containing compounds derived from coal minerals can contribute to excessive gasification of coke in the blast furnace and attack of blast furnace refractories, whereas phosphorus and sulfur from coal minerals can be passed on to the hot metal, thus reducing its quality for steelmaking.

Mineral matter may occur finely dispersed or in discrete partings in coal and is generally grouped according to origin. A certain amount of inorganic matter and trace elements are derived from the original plants. However, the majority is implaced either during the initial stage of coalification (being introduced by wind or water to the peat swamp) or during the second stage of coalification, after consolidation of the coal by movement of solutions in cracks, fissures and cavities. Mineral components of plant origin are not easily recognized in coals because they tend to be disseminated on a submicron level. The primary mineral components incorporated during plant deposition tend to be layered with and intimately intergrown with the organic fraction, whereas the secondary mineral matter tends to be coarsely intergrown and associated with cleat, fractures and cavities. Therefore, secondary minerals may be more readily separated (cleaned or washed) from the organic matrix to improve the value of the material.

d) Industrial Properties

Most of the ancillary mechanical and physical tests used to characterize coals and often included in classification schemes, were developed in support of efforts to identify coals for coke making. As stated earlier, the unique property that sets coking coals apart from other coals, is caking ability. There has been much effort to characterize the swelling, contracting and thermoplastic properties of coals using techniques that allow for the comparison of different coals and how these properties influence coke production and quality. Laboratory tests such as the crucible or free swelling index, Gray-King coke type, Roga Index, Audibert-Arnu Dilatometer and Gieseler Plastometer, provide some means of evaluating the relative strength of swelling, degree of contraction and how fluid a coal will become under heating conditions similar to those encountered during coke making.

Another important mechanical test designed to provide a measure of the ease of pulverization of a coal in comparison with other standard reference coals is the Hardgrove grindability index, (HGI). Ease of grinding is an important economic consideration for all industrial processes. Grindability changes with coal rank, i.e., coals of very low and very high rank are more difficult to grind than middle-rank coking coals. Other factors that influence HGI include the presence of different maceral components, the presence of even small proportions of hard minerals (like quartz) and variations in moisture content. Of these factors, changes in inherent moisture content cause the most variation in the HGI index, particularly for lower rank coals.


a) Coke Making

Coke is produced by heating particulate coals of very specific properties in a refractory oven in the absence of oxygen to about 1100 C (2000 F). As temperature increases inside the coal mass, it melts or becomes plastic, fusing together as devolatilization occurs, and ultimately resolidifies and condenses into particles large enough for blast furnace use. During this process, much of the hydrogen, oxygen, nitrogen, and sulfur are released as volatile by-products, leaving behind a poorly crystalline and porous carbon product. The quality and properties of the resulting coke is inherited from the selected coals, as well as how they are handled and carbonized in coke plant operations.

In terms of coal properties, coke quality is largely influenced by coal rank, composition (reactive and inert macerals and minerals), and an inherent ability when heated to soften, become plastic, and resolidify into a coherent mass. Bituminous class coals of high volatile A, medium volatile, and low volatile rank possess these properties, but not all produce a coke of desirable quality and some even may be detrimental to coke ovens. To compensate for the lack of individual coals with all the necessary properties, blends of anywhere from 2 to 20 different coals are used in today's coke making operations. These coal blends must be managed to optimize coke quality and reduce the cost of raw materials. Individual coals and coal blends need to have the proper proportions of reactive and inert components, must have a relatively low concentration of alkalis-containing minerals, low ash and sulfur yields and be sufficiently thermoplastic to bind all of the components together. At the same time they must provide a level of contraction that will allow the coke mass to be easily removed from a coke oven.

b) Blast Furnace Injection

Carbon from either coke or tuyere-level injectants, such as coal, natural gas, tar or oil can be used in the raceway zone of the blast furnace to generate some of the energy and reducing gases needed to reduce iron oxides, preheat the burden and produce molten metal and slag. In the 1960's, U.S. Steel Corporation began evaluating the technological feasibility for the injection of pulverized coal into the blast furnace as a means of reducing coke rate and hot metal costs. In these early tests it was found that coke rates could be reduced 36% by injecting 230 kg of coal/metric ton of hot metal without enriching the oxygen content of the hot blast. With oxygen enrichment a further reduction of coke rate to 48% could be obtained by injecting 290 kg of coal/ metric ton of hot metal. At the beginning of the 1990's blast furnace injection of coal became the rule rather than the exception as injection rates reached 200 kg of coal/metric ton of hot metal in Europe and Japan.

Generally, coal injection systems consist of a grinding mill that is capable of reducing and drying the 50 x 0 mm (2 x 0 inch) raw coal in one step to a particle-size range acceptable for distribution and injection. Waste gas from the blast furnace can be used in drying operations to reduce cost, but natural gas systems have also been used. Furthermore, an inert atmosphere should be employed to mitigate the potential for fires and explosions of dry coal particles. Two processed coal size specifications are used, a granular system where coal is ground to 95% minus 2 mm (10 mesh) and 20 % minus 0.075 mm (200 mesh), or 100% minus 2 mm and 80% minus 0.075 mm. After crushing, the gas and coal are separated in a cyclone and bag filter system from which the pulverized coal is sent to storage for injection. The coal preparation system represents two thirds of the total capital cost of a coal injection system.

A wide variety of non-coking coals have been tested for injection, ranging from lignite to bituminous coals and anthracites. The choice depends on price and availability rather than attaining the highest injection rates or coke replacement. In light of the diversity of coals that have been used, if all other considerations were equal (raw materials, transportation cost, and availability of tonnage), the level of ash yield, followed closely by volatile matter and moisture content and a coal's grindability are the most important properties.

The yield of ash from a coal should be minimized, but practically speaking should be less than 10%. Ash contributes non-combustible components to the blast furnace burden, has been shown to lower flame temperature and makes grinding difficult. Along with this goes the need to select a coal that contributes a minimal amount of sulfur and alkali content. Volatile matter should be maximized to the extent possible as it has been found that low volatile content coals do not combust completely in the raceway. This is advantageous to reduce possible carryover of particulate materials in the off gas of the blast furnace. However, high volatile content coals generally have higher moisture contents. Moisture should be minimized as additional energy will be needed for its removal in the blast furnace. It also contributes to difficulties during grinding and to problems of free flow from storage bins. Coal hardness, as measured by Hardgrove grindability, should be minimized for the efficient operation of the grinding mills.

c) Alternative Iron Making

Technology Just as improvements in steel making practices by the incorporation of continuous casting technology have eliminated many expensive energy- and labor- intensive steps, similar improvements in the primary end of steel making are being investigated. Direct iron making technologies may permit the use of coal directly without the need for metallurgical coke, may provide lower capital cost compared to the blast furnace/coke oven route, and would have the advantage of being able to offer the economy of small scale production (1,000-2,000 ton/day plants). These techniques may feature high flexibility in raw materials selection.

Although there are variety of direct iron smelting technologies that have been and are being developed, there is only one process that has been in full commercial production since 1989, The Corex process at ISCOR's Pretoria Works, in Republic of South Africa. Other technologies that are currently under development include the American Iron and Steel Institute's Direct Smelting Process, the Australian Hismelt Process, the Russian Romelt Process, Japan's Direct Iron Ore Smelting (DIOS) Process, and the Cyclone Converter Furnace a cooperative effort of Hoogovens and British Steel. Basically, these technologies share similarities in the raw materials employed and smelting procedures. Typically, they use a variety of iron-containing fines, granular coal, oxygen and/or oxygen enriched air for the smelting process. However, differences in smelter operations, the means and methods of blowing air or oxygen through tuyeres at various locations within the smelting vessel, and the use of pre-reduced iron are the primary variables.

In these processes, coal is added directly to the smelting vessel and is the source of reducing gases and thermal energy. Consequently, the least expensive coals of high calorific value that are easily crushed and handled are employed. Thus, non-coking, locally obtained coals of medium volatile through anthracite rank are most often used. Particle size requirements are variable, but minus 1 mm is most often used and, in some cases there is an effort to minimize moisture content to below 6%.

Table 1 - Some of the Parameters Used by ASTM for the Classification of Coals by Rank