EAF Slag Chemistry: Phosphorus, Sulfur, and the Lime-FeO Trade-Off
Slag in an electric arc furnace is not waste. It is the second reaction vessel, and getting its chemistry right is the difference between a heat that meets spec and a heat that goes to the slag pit. The two elements that operators chase hardest are phosphorus and sulfur. They want phosphorus low, sulfur low, and the steel clean. All three goals push on the same slag, and the slag can only do so much at once.
Here is how the trade-offs actually work.
Phosphorus removal is an oxidation reaction. In a basic slag with high CaO and moderate FeO, phosphorus transfers from the metal to the slag as P2O5. The reaction is exothermic, which means it gives off heat - useful during oxygen blowing. The equilibrium is favored by high basicity (high CaO/SiO2 ratio), high FeO in the slag (typically 15 to 25 percent FeO during dephosphorization), and lower temperature. Wait - lower temperature? Yes. Phosphorus removal is one of the few steelmaking reactions that prefers a cooler bath. That is why operators try to remove phosphorus early in the heat, before the bath reaches full temperature.
A typical EAF dephosphorization window sits at 1550 to 1600 degrees C. Push the bath above 1650 and the equilibrium shifts back toward phosphorus in the metal. That is a problem for high-power UHP shops running flat bath temperatures of 1650 to 1700. Operators handle this by adding the dephosphorization lime early, holding a high-FeO slag, and finishing the dephosphorization step before the bath goes to full tap temperature.
Sulfur removal is the opposite. It is a reduction reaction. Sulfur transfers from metal to slag when the slag is low in FeO, high in CaO, and reducing. Add carbon or aluminum to lower the FeO, and the partition ratio shifts. That is why desulfurization happens late in the heat, after the oxidizing oxygen blow, with carbon injection or a reducing slag cover.
The FeO-Lime trade-off is the real headache for operators.
Phosphorus wants high FeO. Sulfur wants low FeO. You cannot have both at once on the same slag. Operators solve this in two ways.
Way one: two-slag practice. After the dephosphorization step, dump the oxidizing slag (a "white slag" or "foamy slag" with 15 to 25 percent FeO) by tilting the furnace or skimming through the slag door. Build a new reducing slag with fresh lime, fluorspar, and carbon or aluminum for desulfurization. The two-slag practice gets you both low phosphorus and low sulfur, but it costs time (15 to 25 minutes per slag change), flux (extra lime, fluorspar, carbon), and refractory (the slag line wears faster with two slag dumps).
Way two: single-slag practice with a compromise. Hold the FeO at an intermediate level - 10 to 15 percent - and accept slightly less efficient phosphorus and sulfur transfer. Most modern EAFs run single-slag because the time and material cost of two-slag is too high. The single-slag slag typically achieves 0.015 to 0.025 percent phosphorus and 0.015 to 0.025 percent sulfur. That is fine for rebar and structurals, marginal for plate grades, and insufficient for special bar quality (SBQ) and forging grades.
The lime basicity ratio (CaO/SiO2) is the next variable. Higher basicity favors both phosphorus and sulfur removal. The cost is more lime consumption, more slag volume, and more refractory wear. A typical EAF single-slag practice runs basicity of 1.8 to 2.5. Two-slag practice can go to 3.0 or higher during the dephosphorization step.
Lime consumption is a real cost line. Burnt lime (CaO) costs $80 to $150 per ton in most markets. Dolomite or dolomitic lime adds MgO, which helps protect the refractory and contributes to basicity. A typical EAF single-slag heat uses 30 to 50 kg of burnt lime per ton of steel, plus 5 to 15 kg of dolomite. That is $3 to $8 per ton in flux cost. Two-slag practice can double that.
Fluorspar (CaF2) is the other flux that operators use sparingly. It lowers slag viscosity, helping the slag flow and react. But it is expensive - $200 to $400 per ton - and it attacks refractory aggressively. Most EAF operators limit fluorspar to 1 to 3 kg per ton, used only when the slag gets too viscous.
Foamy slag is where the chemistry meets the physics.
A foamy slag is a slag with gas bubbles trapped inside it - typically CO from carbon reacting with FeO, plus some air entrained during oxygen blowing. The gas fraction runs 30 to 60 percent of the slag volume. That foam insulates the arc, buries the electrode tips, and protects the sidewall refractory from radiant heat. The gas is generated by injecting carbon - pulverized coal, pet coke, or anthracite - through the oxygen lance or through lances in the slag door.
The carbon injection rate is the operator's main control. Too little carbon and the slag does not foam - the arc flares, hits the sidewall, and the refractory wears fast. Too much carbon and the slag over-foams, slopping over the slag door, and you lose carbon inventory. The sweet spot for most EAFs is 8 to 15 kg of carbon per ton of steel during melt-in and flat bath.
The FeO content of the slag drives the carbon reaction. At 15 to 25 percent FeO, the carbon reacts quickly and the foam builds within 2 to 4 minutes. At 5 to 10 percent FeO, the reaction is slow. At less than 5 percent FeO, there is no foam - the slag sits there as a flat, dense layer.
Bottom stirring changes the picture.
A bottom-stirred EAF uses porous plugs or tuyeres in the hearth to bubble argon (or nitrogen) through the bath. The gas stirs the metal-slag interface, accelerates the slag-metal reactions, and helps float inclusions. Bottom-stirred EAFs achieve 30 to 50 percent better phosphorus and sulfur transfer for a given slag chemistry. The cost is the stirring gas (roughly $1 to $3 per ton) and the maintenance on the porous plugs (replace every 200 to 400 heats).
Modern EAF control systems integrate slag chemistry measurement with bath temperature and carbon content. Some use a sublance that dips into the bath mid-heat, measures temperature and carbon, and pulls a sample for the lab. The lab result drives a slag adjustment - more lime, more carbon, more oxygen - in real time. This level of control is standard on newer EAFs and is being retrofitted on many older ones.
The tapping decision is where slag chemistry meets operational reality.
For a heat running single-slag practice, the operator decides whether to dump the slag at tap or carry it over to the ladle. Carrying slag to the ladle means losing some yield (steel trapped in the dumped slag at the EAF goes to the slag pit) but improves downstream cleanliness because the EAF slag continues to react with the metal in the ladle during ladle furnace (LF) treatment. Dumping the slag at tap means cleaner ladle slag for the LF, but higher EAF yield loss.
Eccentric bottom tap (EBT) systems changed this picture. An EBT furnace pours 95 percent or more of the steel out through a taphole at the bottom of the shell, with the slag retained in the furnace. The operator can dump the slag cleanly through the slag door before tap, or carry it over if the downstream process needs it. Most EBT operations dump before tap, then build a new slag in the ladle using ladle additives.
The bottom line on EAF slag chemistry. Phosphorus and sulfur removal are two reactions that fight each other for control of the slag. Operators pick a side - efficient single-slag, time-consuming two-slag, or compromise with bottom stirring and process control. The lime-FeO balance is the central knob. Get that balance right for your steel grade, your heat time, and your flux budget, and the rest of the slag chemistry follows.
Author: MONTE INTELLIGENCE EAF process team. For slag chemistry optimization studies and EAF training, contact helenxu@cnlymonte.com.
