If alumina is what a refinery buys, reactive silica is what it pays to get rid of. It is one of the most important numbers on a metallurgical bauxite specification, and one of the most often misunderstood.
Not all silica is the same
Silica in bauxite comes in two broad forms.
- Quartz, crystalline SiO₂, is largely unreactive at the temperatures used to digest gibbsitic bauxite. It passes through to the residue.
- Kaolinite and similar clay minerals contain silica that dissolves readily in hot caustic soda. This is reactive silica.
A sample's total silica therefore tells you little on its own. Two bauxites with identical total SiO₂ can have very different reactive silica, and so very different costs in the refinery.
What reactive silica does in the Bayer process
When kaolinite dissolves in Bayer liquor, the silica does not stay in solution. It re-precipitates as a sodium aluminosilicate, often called desilication product, that settles with the residue. Each time it does, it locks away:
- Caustic soda, which the refinery must replace, a major operating cost.
- Alumina, which is lost to the residue instead of becoming product.
It also adds to the volume of red mud and can form scale on heat exchangers and pipes if not controlled. That is why refineries run a desilication step before digestion, and why they write tight limits into bauxite contracts.
How it is measured
Reactive silica is normally determined by a laboratory digestion test that simulates refinery conditions, then measures how much silica reacted. Results depend on the method and temperature, so contracts should specify the procedure. For high-temperature refineries treating boehmitic ore, some quartz also becomes reactive, and the definition shifts accordingly.
Refiners also look at the ratio of available alumina to reactive silica. The higher the ratio, the more product per unit of caustic soda lost.
Where it comes from in a lateritic deposit
In a Konkan-style lateritic profile, reactive silica is concentrated in the clay-rich parts of the deposit, especially towards the lithomarge layer at the base of the ore and in pockets where weathering was incomplete. Thoroughly leached gibbsitic zones are typically lower in clay. See how lateritic bauxite forms.
How a mine keeps it in check
- Sampling before mining. Grid sampling maps where the cleaner ore sits, so benches can be planned around it.
- Selective mining. Keeping the floor of the pit above the clay-rich layer avoids diluting good ore.
- Screening. Clay often reports to the fine fraction, so size separation can lower silica in the product.
- Grading and blending. Lots are sorted by chemistry so each buyer receives material within their limits.
- Testing twice. Checking at the stockpile and again at loading catches drift before it reaches the customer.
These are the same steps described in our pit-to-port operations. Grading lets one ore body serve buyers with different silica tolerances: high-silica material that does not suit a refinery may still be suitable for other uses.