Why FeSi75 Has Such a Strong Effect on Recovery
Recovery efficiency is the share of the alloy that dissolves and reports to the steel, measured against the share lost to oxidation, slag reaction and burn-off. FeSi75 carries a high silicon content, and silicon has a strong affinity for oxygen. That single fact explains why the same product can return a very high recovery in one heat and a poor recovery in the next: both the wanted deoxidation reaction and the unwanted oxidation reaction are driven by the same chemistry.
When FeSi75 is added under controlled conditions, most of the silicon dissolves and mixes into the bath before it can react with oxygen at the slag-metal interface. When the addition is made into a turbulent bath, under a thin or highly oxidising slag, or with the alloy held on the surface for a long time, a larger fraction of the silicon is consumed as a deoxidiser and floats out in the slag instead of alloying the steel. The alloy chemistry sets the ceiling on recovery; the addition practice decides how close to that ceiling a heat actually runs.
How the Size Fraction Changes Dissolution Behaviour
Particle size controls the surface area available for reaction, the melting time and the distribution of the alloy through the bath. The 10–50 mm band is the most widely traded export fraction because it offers a balanced melting profile, and the comparison below explains why.
| Size band | Behaviour in molten steel | Effect on recovery |
|---|---|---|
| 3–10 mm | Fast dissolution, large exposed surface area | Lower recovery unless feeding rate and slag condition are tightly controlled |
| 10–50 mm | Balanced melting with steady, predictable dissolution | Generally the highest and most consistent silicon recovery |
| 50–100 mm | Slow melting, risk of unmelted remnants | Lower and more variable recovery, particularly in small heats |
Fine fractions melt quickly, which is useful when a fast chemistry correction is needed, but the same surface area that speeds melting also accelerates oxidation before the silicon can be mixed into the bath. Coarse lumps melt slowly, and if the bath cools while they are still dissolving or the stirring is weak, part of the addition never reaches the bulk liquid and is lost to the slag on the next slag-off. The 10–50 mm band reduces both failure modes, which is why it has become the standard export specification.
Size distribution within a lot matters as much as the nominal range. A lot that certifies as 10–50 mm but carries a heavy fines tail behaves closer to a fine fraction in the furnace, because the fines react first and set up a local high-oxygen zone around the addition point.
Impurity Control and Its Effect on Recovery
The impurity profile of FeSi75 is not only a chemistry question; it changes the way the alloy behaves while it dissolves. Aluminium, calcium, sulphur and carbon each influence melting dynamics and slag-metal reactions.
Aluminium is the most important of the four. A higher aluminium level increases oxide formation in the addition zone, diverting part of the silicon into inclusions instead of into solution, and it also contributes aluminium to the steel chemistry whether or not the heat was intended to receive it. Calcium and sulphur change slag fluidity and therefore the time window in which the silicon can dissolve before the slag is tapped or the heat is treated. Carbon carried in the alloy adds to the carbon pickup calculation in low-carbon and ultra-low-carbon grades.
Ferrosilicon grades and tolerances are published nationally, with silicon content as the primary basis and aluminium as the sub-grade separator. The practical requirement for a recovery-focused buying specification is threefold: a silicon range, a maximum aluminium figure, and the size fraction, all quoted against the same standard number on the purchase order so that supplier and inspector are working from one document.
| Element | Mechanism | Practical consequence |
|---|---|---|
| Aluminium | Forms oxides in the addition zone | Inclusion load increases, silicon recovery falls, aluminium reports to the steel |
| Calcium | Alters slag fluidity and reaction time | Shifts the window in which dissolution must be completed |
| Sulphur | Interacts with slag and desulphurisation practice | Affects the slag condition during and after the addition |
| Carbon | Adds directly to the bath carbon balance | Must be counted in low-carbon and ultra-low-carbon heats |
Addition Practice: The Largest Single Variable
A consistent alloy cannot compensate for an inconsistent addition. The variables that matter most are where the alloy is added, when in the refining sequence it is added, and how well the bath is mixed at the time of addition.
Adding FeSi75 too early exposes the alloy to a bath that is still highly oxidising and to a slag that has not yet been conditioned. Adding it too late leaves insufficient time for mixing, so a portion of the silicon remains undissolved when the heat is tapped. Both errors raise the apparent consumption per heat even when the alloy itself is perfectly within specification.
Good stirring and a stable slag cover support recovery by minimising the exposure of freshly exposed alloy surfaces to oxygen. In practice this means adding into a stirred bath at a controlled rate, avoiding dumping the full quantity into one spot, and maintaining enough slag viscosity that the alloy sinks through it rather than floating on it.
Timing relative to temperature is equally important. Silicon dissolves more readily and oxidises less when the addition is aligned with a stable, adequately superheated bath. Heats that are treated cold show lower recovery and require a second correction, which then compounds the loss.
Energy, Consumption and Cost Effects of Better Recovery
Recovery has a direct cost consequence that is often larger than the price difference between two suppliers. When recovery is high, the plant reaches the target silicon level without overfeeding. That reduces total alloy consumption, shortens the correction sequence, and stabilises the chemical adjustment so that reblows become less frequent. Over a full campaign the effect is visible in the consumption figure per tonne of steel rather than in any single heat.
Where recovery is unstable, the usual response is to feed a safety allowance above the calculated requirement. That allowance is the most expensive material in the heat, because it is the portion of the alloy most likely to be lost to the slag. Measuring recovery consistently, heat by heat, and correcting the allowance from the measured average rather than from a fixed margin is the quickest route to lower consumption.
Estimating recovery requires three numbers: the mass of alloy added, the analysed chemical content of that alloy, and the analysed silicon content of the steel sampled after mixing. Without an alloy sample per lot, the third number cannot be closed and any recovery figure remains an assumption.
Frequently Asked Questions
Q: Why does FeSi75 affect recovery more than a lower-grade ferrosilicon?
A: FeSi75 carries a higher silicon content per tonne, so a fixed oxidation loss removes a larger absolute quantity of the element being paid for. That concentrates both the benefit of good practice and the cost of bad practice into the same addition.
Q: Is 10–50 mm really better than 3–10 mm for recovery?
A: For most secondary steelmaking routes yes, because the balanced melting profile limits the exposure of fresh alloy surface to oxygen while still completing dissolution within the treatment window. Fine fractions are appropriate where a rapid correction is required and slag conditions are tightly controlled.
Q: Does a higher aluminium content in the alloy lower silicon recovery?
A: Aluminium oxidises readily in the addition zone and increases the local oxide load, which diverts silicon into inclusions. It also reports to the steel aluminium content, so it must be counted in grades with a tight aluminium specification.
Q: When should FeSi75 be added in the refining sequence?
A: After the slag has been conditioned and the bath is stirred, adequately superheated, and past its most strongly oxidising stage. Adding too early wastes alloy to oxidation, and adding too late leaves unmelted material at tap.
Q: How is recovery actually measured?
A: From the mass of alloy added, the analysed silicon content of the alloy, and the analysed silicon content of the steel sampled after complete mixing, all expressed on the same weight basis. Recovery calculated without an alloy analysis per lot is an estimate, not a measurement.
Q: Can recovery be improved without changing the alloy supplier?
A: Yes. Addition timing, feeding rate, stirring condition and slag control are all operational variables, and in most plants they account for a larger change in recovery than the chemistry difference between compliant suppliers.
