Basic oxygen steelmaking is a primary metallurgical process that converts carbon-rich molten iron into steel by injecting high-purity oxygen. Oxidation removes carbon and other dissolved elements while releasing enough heat to keep the metal molten and melt a substantial quantity of steel scrap. The reactions take place in a large, tiltable, refractory-lined vessel known as a basic oxygen furnace or converter. “Basic” refers to the chemical character of its lining and lime-rich slag, not to the complexity of the process.
Developed commercially in Austria in the early 1950s, the process became the principal steelmaking method associated with integrated iron and steel works. It combines large batch sizes, short processing times, and effective control of composition. Its role is best understood within the complete production chain: a blast furnace supplies molten iron, the oxygen converter performs rapid primary refining, and subsequent ladle treatments prepare the steel for casting. Modern practice encompasses top-blown, bottom-blown, and combined-blowing arrangements, supported by extensive gas cleaning, material recycling, and computerized process control.
Origins and industrial adoption
Basic oxygen steelmaking developed from an older principle: impurities dissolved in molten iron can serve as fuel when oxidized. Henry Bessemer demonstrated the industrial importance of this principle in the 1850s with the Bessemer process, which blew air through molten iron. The original acid-lined converter could not effectively remove phosphorus. Sidney Gilchrist Thomas and Percy Gilchrist subsequently developed a basic-lined version in the late 1870s, allowing phosphorus-rich iron to be refined with a lime-bearing slag. These processes established the importance of converter geometry, refractory chemistry, and rapid oxidation.
Air blowing nevertheless imposed significant limitations. Nitrogen, which makes up most of the air, absorbed heat and entered the steel to varying degrees. Replacing air with oxygen promised greater thermal efficiency and lower nitrogen pickup, but early oxygen experiments encountered practical difficulties, including intense localized temperatures and rapid damage to bottom-blowing components. Large-scale oxygen production and effective delivery equipment were therefore prerequisites for a commercially successful process.
An important step came through Robert Durrer’s work in Switzerland in the 1940s, which demonstrated the possibilities of blowing oxygen onto the surface of molten iron. Austrian steelmakers subsequently developed the method for large-scale production. Commercial operation began at Linz in 1952 and Donawitz in 1953. The designation “LD,” derived from these two locations, became widely associated with the top-blown process. A water-cooled lance delivered oxygen from above, avoiding the severe refractory problems associated with injecting unprotected oxygen directly through the converter bottom.
The process spread rapidly during the following decades, displacing much open-hearth steelmaking. An open-hearth furnace typically required several hours to refine a heat, whereas an oxygen converter could complete its blowing stage in roughly 15–20 minutes. Its success depended not only on faster refining but also on the availability of large oxygen plants, reliable refractory materials, suitable cranes and charging equipment, and continuous supplies of molten iron.
Later development focused on improving bath mixing, endpoint accuracy, vessel life, and environmental performance rather than changing the central oxidation principle. Bottom oxygen injection using protected tuyeres became commercially important around the end of the 1960s and the early 1970s. Combined-blowing systems subsequently added bottom stirring to top-blown converters. These developments produced a family of oxygen-converter practices rather than a single fixed furnace design.
Metallic charge, fluxes, and oxygen supply
The principal feedstock is hot metal from a blast furnace, commonly containing about 4–4.5% carbon by mass, together with silicon, manganese, phosphorus, and sulfur. Its temperature on arrival at the steelmaking shop is often approximately 1,250–1,400 °C, depending on transport and upstream operations. Both its sensible heat and its oxidizable elements contribute to the converter’s heat supply. Consequently, hot-metal composition is a major determinant of how much scrap can be melted and how much slag must be produced.
Steel scrap commonly accounts for roughly 15–30% of the metallic charge in conventional practice, although the actual proportion varies substantially. Scrap cools the process as it heats and melts, while also recovering previously produced iron. Heavy pieces, light sheet scrap, and prepared bundles differ in bulk density, charging behavior, and melting rate. Oversized or poorly distributed scrap can remain unmelted or interfere with refining. Copper and tin introduced with scrap are especially important because ordinary oxygen-converter refining does not remove them effectively.
Other iron-bearing materials can supplement the charge. Iron ore, mill scale, and certain recycled oxide materials supply iron while acting as coolants through heating and reduction. Direct reduced iron or hot-briquetted iron may also be used under suitable conditions. Their metallization, residual oxide content, gangue, and temperature affect their thermal value. An oxide-rich addition cannot be treated as equivalent to an equal mass of metallic scrap because reducing its iron oxides consumes carbon and heat.
The principal fluxes are lime, mainly calcium oxide, and dolomitic lime, which also supplies magnesium oxide. Lime combines with silica and helps establish conditions for phosphorus removal. Magnesium oxide additions help control slag attack on the refractory lining. Flux performance depends on particle size, reactivity, and dissolution rate: a calculated amount of lime is ineffective if much of it remains undissolved during the period when refining requires a liquid, chemically active slag.
Oxygen is generally produced in a nearby air-separation plant, usually by cryogenic distillation. Converter oxygen commonly exceeds 99% purity. Its production, compression, storage, and distribution must support a short interval of very high demand. Consumption is often of the order of 50–60 normal cubic metres per tonne of liquid steel, although charge chemistry, iron oxidation, gas reactions, and operating practice change this figure. A normal cubic metre denotes gas volume at specified reference conditions; comparisons require the same temperature and pressure convention.
Converter construction and supporting equipment
A basic oxygen converter consists of a welded steel shell lined with heat-resistant brick and supported on trunnions that allow it to tilt. The shell is generally pear-shaped, with a relatively narrow mouth above a wider reaction chamber. Tilting permits scrap and hot-metal charging, sampling, steel tapping, and slag discharge. Industrial vessels vary widely in size; many large converters produce approximately 150–350 tonnes of steel per heat, while smaller and larger installations also exist. Their working volume must accommodate vigorous gas evolution and slag foaming above the molten bath.
The working lining commonly uses magnesia–carbon refractories. Magnesium oxide resists basic slags, while carbon improves resistance to slag penetration and thermal shock under appropriate conditions. Different regions experience different wear mechanisms: impact during charging, chemical dissolution near the slag line, erosion around the tapping hole, and localized damage near bottom stirring elements. Brick quality and thickness are therefore zoned rather than necessarily uniform throughout the vessel.
In a top-blown converter, oxygen enters through a vertically movable, water-cooled lance. Its copper tip contains several nozzles arranged to distribute jets across the bath. The nozzles are designed for high-velocity flow, often supersonic under operating conditions. Lance height, oxygen pressure, nozzle geometry, and bath depth determine jet penetration and the balance between surface interaction and metal ejection. Too little penetration can produce an excessively oxidized slag, whereas overly aggressive interaction can increase splashing and refractory wear.
Water cooling protects the lance but introduces a critical hazard if water leaks into molten metal. Flow, pressure, and temperature monitoring are therefore integral to its operation. The lance mechanism must also withdraw the equipment rapidly when abnormal conditions arise. Some installations use specialized lance designs or additional oxygen outlets to influence post-combustion or slag behavior, but the thermal benefit depends on where the released heat is transferred.
Above the vessel, a hood captures hot, dust-laden gas. Cooling equipment, ducts, gas-cleaning systems, fans, and sometimes gas-storage facilities form a major part of the installation. The converter shop also contains charging cranes, scrap boxes, hot-metal ladles, flux bins, alloy systems, and steel ladles. A tap hole in the vessel wall allows steel to be poured separately from most of the slag. This separation is imperfect, making slag-detection and slag-stopping equipment important links between primary refining and later ladle treatment.
Oxidation reactions and slag chemistry
The essential transformation is decarburization: dissolved carbon reacts with oxygen and leaves principally as carbon monoxide. A simplified reaction is
Square brackets conventionally denote an element dissolved in the metal. Some carbon monoxide subsequently burns to carbon dioxide, either within the vessel’s upper region or in the gas-handling system. The location of this additional combustion matters because only part of the heat may return to the bath; the remainder can heat the outgoing gas or surrounding equipment.
Silicon, manganese, iron, and phosphorus also oxidize. Silicon oxidation produces silica, which consumes lime in forming the slag. Manganese partitions between metal and slag according to temperature and oxygen potential. Iron oxidation produces iron oxides that help form a fluid, reactive slag but represent a loss of metallic yield if not subsequently reduced. The converter therefore cannot be understood simply as oxygen reacting selectively with carbon. It is a coupled system involving metal, slag, gas, and suspended droplets.
Phosphorus removal requires an oxidizing, sufficiently basic slag. A simplified representation is
where parentheses indicate slag components. This equation represents phosphorus transfer and stabilization rather than the complete molecular structure of an industrial slag. High lime activity, suitable iron-oxide content, and relatively low temperature generally favor phosphorus retention. Silicon-rich hot metal increases the lime requirement because silica must first be accommodated. Inadequate slag separation can later allow phosphorus to return to the steel under reducing conditions.
Slag basicity is often expressed as the mass ratio
This ratio is useful but incomplete. Magnesium oxide, alumina, phosphorus oxides, iron oxides, temperature, and the amount of undissolved lime all influence slag properties. Two slags with the same nominal basicity can differ substantially in viscosity, liquid fraction, and refining ability. Excessive additions can increase slag volume and heat demand without delivering a proportional improvement in impurity removal.
Reaction rates depend on mass transfer as well as thermodynamics. Oxygen jets eject metal droplets into the slag, creating a large interfacial area. Carbon monoxide generated around reacting droplets contributes to a foaming slag–metal–gas emulsion. This emulsion accounts for much of the converter’s rapid refining. Sulfur behaves differently from phosphorus: deep desulfurization is favored by low oxygen activity and basic slags, so it is generally more effective during hot-metal pretreatment or reducing ladle treatment than during the strongly oxidizing converter blow.
The operating cycle of a heat
A converter heat begins with a charge calculation based on the required steel grade, target tapping temperature, hot-metal analysis, scrap inventory, and expected process losses. The calculation must satisfy both mass and energy balances. Carbon, silicon, iron, and other elements are distributed among steel, slag, gas, and dust, while heat released by oxidation must cover melting, heating, chemical reactions, and losses. Uncertainty in hot-metal temperature or scrap composition can affect the final result even when weighing equipment performs correctly.
The vessel is tilted for charging. Scrap is commonly introduced first, followed by molten iron, although detailed sequences differ between plants. Charging produces substantial dust and fume, requiring secondary extraction as well as the main converter hood. Once the converter is upright, the oxygen lance is lowered and blowing begins. Lime and other additions are introduced according to the selected practice, often in stages to promote early slag formation without excessive cooling or accumulation of undissolved material.
During the blow, silicon is oxidized rapidly and a slag develops. Carbon removal accelerates as the bath becomes vigorously agitated and a reactive emulsion forms. In the main decarburization period, the rate is strongly related to oxygen delivery. Near the endpoint, as dissolved carbon becomes scarce, transport of carbon to reaction sites becomes increasingly important and a larger share of oxygen may oxidize iron. This shift makes excessive blowing costly in both yield and refractory exposure.
The desired endpoint combines several conditions rather than a single carbon value. Carbon and phosphorus must meet the grade requirements, the steel must be hot enough for downstream processing, and slag oxidation must remain manageable. Tapping temperatures are commonly around 1,600–1,700 °C, depending on composition, ladle condition, and subsequent treatment. A sample and temperature measurement may confirm the endpoint. If necessary, a short corrective blow or coolant addition is made, although additional interventions lengthen the cycle and can increase losses.
The converter is then tilted to tap steel into a ladle, generally with deoxidizers and alloys added during or after tapping. Most slag is retained and discharged separately. The vessel may undergo inspection or lining maintenance before the next charge. A complete tap-to-tap cycle often takes approximately 30–50 minutes, substantially longer than the oxygen-blowing interval because it includes handling, measurement, tapping, and turnaround. Converter-shop capacity consequently depends on coordinated cranes, ladles, casting schedules, and maintenance, not merely the maximum oxygen flow.
Top blowing, bottom blowing, and combined practice
The original LD arrangement supplies refining oxygen through a lance above the bath. This configuration keeps the principal oxygen-delivery equipment outside the molten metal and permits rapid lance replacement. The jets generate localized reaction zones and eject droplets that participate in emulsion refining. However, bath circulation is not uniform, and top blowing alone may leave composition or temperature differences between regions, particularly near the end of the blow when carbon monoxide generation becomes less vigorous.
Bottom-blown oxygen converters inject oxygen through tuyeres below the bath surface. Commercial systems such as OBM and Q-BOP use arrangements that protect the tuyere region, notably by introducing a hydrocarbon coolant around the oxygen stream. Hydrocarbon decomposition absorbs heat and helps prevent destructive local overheating. Because gas enters directly beneath the metal, bottom blowing produces strong circulation and promotes contact between dissolved carbon and the oxidizing phase.
These advantages come with different maintenance and chemical considerations. Bottom components are exposed to severe service, and their condition can constrain operating campaigns. Hydrocarbon injection affects the gas balance and can introduce hydrogen, which must be considered in relation to steel quality and later treatment. Bottom-blown converters therefore require their own operating practices rather than being interchangeable with top-blown vessels.
Combined blowing retains the top oxygen lance while introducing stirring gas through bottom plugs or tuyeres. Argon or nitrogen may be used, with gas selection depending on steel grade, process stage, and cost. Nitrogen is less expensive but can be absorbed by the steel; argon avoids that particular source of nitrogen pickup. Some designs also introduce oxygen through bottom elements. “Combined blowing” thus describes a broad category rather than one standardized nozzle layout.
Bottom stirring improves homogenization and can bring the metal and slag closer to equilibrium. At comparable endpoint carbon, effective stirring can reduce the need for excessive iron oxidation and improve the consistency of phosphorus removal. Its effectiveness depends on gas flow, plug distribution, bath depth, and the condition of the stirring elements. Excessive stirring can increase splashing or wear, so maximum flow is not automatically optimal.
Plants also adapt blowing practice to difficult feedstocks. High-phosphorus hot metal may justify pretreatment, intermediate slag removal, or a double-slag process in which an early phosphorus-rich slag is discarded before refining continues. Such procedures improve impurity control but add handling, heat loss, and cycle time. Vessel configuration and operating sequence are therefore selected together with the available hot-metal chemistry and required product range.
Measurement, modeling, and endpoint control
Converter refining presents a difficult process-control problem because most reactions occur inside an opaque, intensely agitated vessel. Continuous direct measurement of bath composition is generally unavailable. Operators and control systems instead combine initial charge data, intermittent metal measurements, gas analysis, and physical models. The central task is to infer the evolving condition of the bath early enough to achieve the target composition and temperature without repeated corrective operations.
Static models calculate charge requirements before blowing. Their inputs commonly include hot-metal mass, temperature and analysis, scrap type, flux quantities, oxygen purity, and desired endpoint conditions. The models estimate oxygen demand, coolant requirements, slag formation, and heat loss. They are usually calibrated to plant data because important quantities—such as effective heat losses, scrap melting behavior, and iron carried into dust—cannot be represented perfectly from first principles in routine production.
Dynamic control updates the prediction during the blow. Off-gas analysis measures carbon monoxide, carbon dioxide, and sometimes oxygen and other species. Combined with gas-flow measurements, these data provide information about carbon removal and air ingress. In principle, integrating carbon leaving in the gas helps estimate remaining bath carbon, but delays, leakage, dust, and uncertain gas flow complicate the calculation. Signals must therefore be interpreted within a consistent material balance.
A sublance can enter the vessel during refining to obtain temperature, samples, or other measurements without interrupting production as much as conventional vessel tilting. Depending on equipment, probes may also measure dissolved oxygen or provide information related to carbon content. These observations allow the control system to correct its estimate before the endpoint. Optical and acoustic methods can supplement direct probes by tracking flame characteristics, vibration, or signs of excessive foaming.
Modern systems integrate these measurements through automation and statistical models. Machine-learning methods can identify relationships between operating history and outcomes such as endpoint accuracy, slopping, or refractory wear. Their reliability depends on representative data, stable measurement systems, and explicit treatment of changing raw materials and equipment condition. A model trained on one converter campaign or scrap mix may lose accuracy when those conditions change.
Control objectives also compete. A practice that maximizes phosphorus removal may increase slag volume or iron oxidation; one that shortens blowing may reduce the margin for complete scrap melting. Effective endpoint control therefore balances composition, temperature, yield, cycle time, and equipment constraints. It is not simply a matter of stopping oxygen flow when one estimated concentration reaches its target.
Steel quality and downstream refining
The converter performs bulk refining, but it does not normally deliver the final chemical and cleanliness specification on its own. At tapping, the steel contains dissolved oxygen and may carry entrained slag or oxide particles. Secondary metallurgy in the ladle adjusts composition and temperature, removes selected impurities and gases, and controls nonmetallic inclusions. This division of work allows the converter to operate rapidly while more precise finishing occurs in equipment designed for quieter and more controlled treatment.
Deoxidation is a central step. Aluminium, silicon, or manganese may be added to react with dissolved oxygen. Aluminium deoxidation can be represented as
The resulting oxide is not automatically harmless: solid alumina particles can form clusters, affect product properties, or contribute to nozzle clogging during casting. Ladle stirring promotes contact between inclusions and slag, while suitable slag chemistry helps absorb them. Calcium treatment is used for some grades to modify inclusion composition and behavior, but its effectiveness depends on oxygen, sulfur, temperature, and addition practice.
Slag carryover from the converter can impair these operations. An iron-oxide-rich slag consumes deoxidizers and alloy additions, lowering recovery and generating additional inclusions. Phosphorus retained in the slag can also revert to the metal when conditions become reducing. Slag-stopping devices, tap-stream monitoring, and carefully controlled tapping therefore have consequences beyond the converter’s immediate metallic yield.
Gas control is particularly important for demanding products. High-purity oxygen limits the nitrogen introduced with the refining gas compared with air-blown processes, but nitrogen can still enter through air contact, scrap, additions, or bottom stirring. Vacuum treatment, including circulating-vacuum processes used for some flat-product grades, can remove hydrogen and support further decarburization. Ultra-low-carbon steels used in applications such as automotive sheet generally require treatment beyond ordinary converter endpoint control.
Some residual elements remain difficult to remove throughout conventional steelmaking. Copper and tin are not readily oxidized out of molten steel under ordinary converter conditions, so their concentrations are controlled mainly through charge selection and dilution. By contrast, carbon, silicon, manganese, and phosphorus can be substantially altered during oxygen refining. This distinction explains why a high scrap fraction must be assessed in terms of both thermal feasibility and residual-element quality.
After ladle treatment, steel is commonly solidified by continuous casting. Converter timing, ladle availability, and caster sequence must be coordinated to avoid excessive holding or temperature loss. The finished properties then depend on solidification, rolling, heat treatment, and composition; the converter establishes essential starting conditions but does not alone determine the properties of the final product.
Yield, energy recovery, emissions, and safety
Converter efficiency is measured through more than tonnes produced per hour. Metallic yield reflects how much charged iron reaches the tapped steel rather than leaving in slag, dust, spills, or retained material. Some apparent mass loss is unavoidable because carbon and other impurities are deliberately removed. Excessive iron oxidation, however, is a process loss. Slag practice, endpoint accuracy, splash control, and recovery of iron-bearing by-products all influence the amount of raw material needed per tonne of steel.
Carbon monoxide in converter gas contains recoverable chemical energy. In suppressed-combustion systems, air ingress is limited and suitable gas is cleaned and collected for use as fuel elsewhere in the works. Gas quality changes during the blow, so collection is normally restricted to appropriate operating periods. Other installations burn more of the gas near the hood and may recover sensible heat through steam generation. The balance between fuel recovery, steam production, and direct heat transfer depends on plant design.
Dust and fume arise from oxidation, metal ejection, condensation, and material handling. Primary extraction captures gas from blowing, while secondary extraction serves charging, tapping, and other open operations. Wet scrubbers or dry gas-cleaning equipment remove particulate matter. Dust and sludge can contain substantial iron, but recycling is affected by zinc and other contaminants that may accumulate in integrated works. Water treatment and closed-loop cooling reduce discharges without eliminating the need to manage concentrated residues.
Converter slag can be processed to recover entrained metal and, where its properties and local specifications permit, used in construction or other applications. Its suitability is not automatic. Free lime and magnesium oxide can hydrate and expand, while composition and leaching behavior require assessment. Aging, stabilization, grading, and quality control may therefore be necessary before beneficial use.
The process also presents severe process-safety hazards. Water entering molten metal can cause a violent steam explosion, making dry scrap, sound cooling circuits, and controlled equipment condition essential operating requirements. Carbon monoxide is toxic and combustible; gas systems require monitoring, controlled isolation, and measures against explosive mixtures. Slopping—the uncontrolled overflow of foaming slag and metal—can expose personnel and equipment to hot material and disrupt gas capture. Charging impacts, vessel movement, oxygen service, and heavy ladle handling add mechanical and fire hazards.
These risks are addressed through engineered containment, interlocks, gas detection, equipment monitoring, maintenance, and separation of personnel from hazardous operating zones. Process models may help anticipate slopping or abnormal conditions, but they supplement rather than replace physical safeguards and independently functioning protection systems.
Industrial role and the transition in steel production
Basic oxygen steelmaking is closely associated with the integrated blast-furnace route. The blast furnace continuously produces molten iron, while converters refine it in batches. High throughput and short conversion times make this arrangement effective where large volumes of primary iron are required. Its economic performance depends on the entire works: ore preparation, coke supply, ironmaking, oxygen production, energy integration, casting, and the product mix. The converter itself is only one part of the capital-intensive system.
Its principal alternative is electric arc furnace steelmaking, which supplies heat electrically and can melt scrap, direct reduced iron, or mixtures that sometimes include hot metal. The distinction is not simply between new and recycled steel. Oxygen converters recycle substantial scrap quantities, while electric furnaces may use large amounts of primary iron to meet quality requirements or compensate for scrap availability. Relative costs depend on electricity, fuels, ore, scrap, metallic quality, and existing infrastructure.
The oxygen converter’s scrap intake is constrained by its heat balance. Increasing scrap reduces the amount of hot metal needed, but eventually the oxidation heat and incoming sensible heat are insufficient to melt the charge and reach tapping temperature. Hot-metal temperature, silicon content, scrap preheating, auxiliary energy, and post-combustion can alter this limit. Each intervention introduces trade-offs involving equipment, emissions, yield, and operating stability. Higher scrap use is therefore a process-integration problem rather than a simple substitution by mass.
The route also has an important relationship with climate change. Much of the greenhouse-gas burden associated with conventional integrated steel production arises upstream, especially from carbon-based iron-ore reduction and coke production. Converter operation additionally generates carbon-containing gas and requires oxygen, fluxes, and utilities. Comparisons must distinguish direct converter emissions from emissions across the complete steelmaking route; excluding blast-furnace iron production would omit a major source.
Decarbonization strategies consequently extend beyond the converter. They include improved material efficiency, increased use of suitable scrap, changes in ironmaking, low-carbon electricity, and carbon capture and storage. Hydrogen-based direct reduction paired with electric melting changes the production chain more fundamentally by reducing reliance on carbon-rich blast-furnace hot metal. Whether an existing converter remains useful depends on the form, temperature, and chemistry of the replacement iron supply.
The pace of change varies with regional energy resources, scrap markets, ore quality, product requirements, and the remaining life of installed assets. Converter improvements can reduce losses and energy demand within existing integrated works, while a transition to a different ironmaking route may require new melting and refining equipment. These physical and economic constraints shape investment decisions at the level of whole steelworks rather than individual furnaces.