Silica Gel Formation in Acid Leaching: 7 Causes and Fixes

Last updated: 2026-11-02

Silica gel formation — not total dissolved silica — is behind roughly 60-80% of chronic filtration failures in acid leach circuits treating clay-rich, saprolitic, or silicate-gangue ores. This guide explains the polymerization chemistry, lists seven causes with symptoms and fixes, and shows how to restore filtration rate before gelation forces a shutdown of the leaching circuit.

Key Takeaways

• Gelation is triggered by supersaturation of monosilicic acid, not by total dissolved silica.
• The highest-risk pH band sits near 2-3, where amorphous silica is least soluble.
• Filter flux collapse, cloth blinding, and thickener overflow turbidity usually appear together.
• Fast acid addition, over-grinding, and uncontrolled hot-liquor cooling are the most common causes.
• Desliming at a 20-40 micron cut plus staged acid addition fixes most plants within one campaign.

What Is Silica Gel Formation in Acid Leach Circuits?

Silica gel formation is a polymerization process that converts dissolved monosilicic acid into an interconnected solid network, and it is the most expensive chemistry mistake an acid leach circuit can make with silicate-bearing ore. The sequence starts innocently. Acid attack on silicate minerals releases silicon as orthosilicic acid, H4SiO4, a small neutral molecule that is fully soluble and, at the monomer stage, harmless: filtration, washing, and clarification all work normally on liquor carrying only monomer.

Monosilicic acid is a weak acid that exists almost entirely as the neutral H4SiO4 species below roughly pH 9, which is why it slips past the free-acid titrations and pH probes that operators rely on for routine leach control. The trouble begins when concentration crosses saturation. Amorphous silica dissolves to roughly 100-140 mg/L as SiO2 at 25 °C (indicative), and that solubility rises with temperature. Above saturation, monomer condenses with monomer: two H4SiO4 molecules release one water molecule and form a siloxane (Si-O-Si) dimer, dimers build oligomers, oligomers build spherical polysilicic particles a few nanometers across, and if the particles keep growing and colliding they link into a continuous gel that immobilizes the entire liquor volume. Each step accelerates the next, so polymerizing liquor is a running reaction, not a static solution.

Supersaturation, not total silica, is the trigger. A liquor holding 80 mg/L dissolved SiO2 against a 100-140 mg/L equilibrium is indefinitely stable; push the same liquor toward 250-400 mg/L (supersaturation ratios of 2-4×, indicative) and it becomes a clock ticking toward gelation, with the rate set by the supersaturation ratio, pH, temperature, ionic strength, and available nucleation surfaces. That is why two plants leaching ores with identical head silica can live in different worlds: the risk sits in how fast each circuit adds dissolved silica relative to the solubility ceiling, not in how much silica the ore contains.

Where Does the Silica Come From and When Does It Turn to Gel?

Acid leaching does not create silica; it liberates what the gangue already holds. The contributors, in rough order of reactivity in sulfuric acid, are:

• Clay minerals (kaolinite, halloysite, smectites) — the dominant source in clay-rich copper oxide and deeply weathered feeds; their layered aluminosilicate structure hydrolyzes readily in hot acid and releases monosilicic acid within minutes to hours (indicative).
• Serpentine and lizardite in saprolitic nickel laterite — magnesium-silicate sheets that dissolve in hot acid and hand over both magnesia and silica at nearly the same rate, which is why gelation is a defining constraint in atmospheric laterite leaching (see our nickel processing applications).
• Talc — slower to attack, but a sustained silica source over long residence times.
• Feldspars — moderate dissolution in hot, strongly acidic liquor.
• Volcanic glass (tuff, ash, obsidian fragments) — the fastest-releasing of all; some glass-bearing ores can gel liquor within a single bottle roll.
• Metal silicates in the ore value minerals — chrysocolla in copper oxide and hemimorphite in zinc silicate ores dissolve with the metal and leave their silica behind in solution.

Quartz is usually the innocent bystander: its dissolution rate in sulfuric acid at normal leach temperatures is far too slow to feed dissolved silica on process timescales. Total SiO2 head grade therefore correlates poorly with gelation risk — a 5% SiO2 ore made of reactive clays is more dangerous than a 30% SiO2 ore made of quartz. Feed blending on head grade alone fails as a gelation control; mineralogy should drive blend decisions.

Solubility behavior sets the timing of the risk. Amorphous silica solubility is strongly pH dependent, with a broad minimum near pH 2-3 where dissolved silica sits closest to saturation for any given release rate — and where gelation risk consequently peaks. The gelation window is a pH range that spans roughly pH 2-6 (indicative), inside which polymerization kinetics accelerate sharply. Most plant excursions — neutralization events, dilution with weak acid, blend swings that change acid demand — drag liquor from comfortable pH <1 conditions into that window while still carrying a high silica load.

What Does Silica Gelation Do to the Downstream Circuit?

Silica gelation rarely announces itself in the leach tanks first; it announces itself in the solid-liquid separation plant, where the costs concentrate. The failure chain is predictable, and most plants report several links at once:

Viscosity rise. Polysilicic acid raises apparent liquor viscosity, sometimes by a factor of 1.5-3 over baseline (indicative), before any visible gel appears. Pumps run further up their curve, agitator power draw creeps, and heat-transfer coefficients fall.

Filtration rate collapse. Colloidal polysilicic particles are exactly the wrong size for a filter cloth: too large to pass, too fine to form a permeable cake. They plug cloth pores and smear across the cake face, so a circuit that filtered comfortably at design rate can drop to a fraction of that flux within days.

Cloth blinding. Silica gel bonds tenaciously to polypropylene cloth, and normal wash cycles recover little of the loss once blinding is established. The fix shifts from operation to mechanical cleaning or cloth replacement.

Thickener overflow turbidity. In counter-current decantation, fine silica colloids settle poorly, carry forward with the overflow, and contaminate the liquor sent to metal recovery; sizing checks belong in the thickener sizing review whenever turbidity trends upward.

High cake moisture. A gel-fouled cake has a collapsed, impermeable structure that holds entrained liquor; wash displacement suffers and residual moisture climbs, with implications for transport, handling, and downstream drying covered in our filter cake moisture guide.

Silica-gypsum scale in CCD and heat exchangers. In sulfate circuits, silica and gypsum co-deposit on CCD launders, interstage piping, and heat-exchanger tubes wherever temperature and pH shift locally, and the wash efficiency assumptions behind your CCD wash ratio design quietly stop being true.

Resin and carbon fouling. Silica polymerizes inside resin beads and carbon pores during loading and especially during regeneration, blocking exchange sites and mimicking a slow, irreversible capacity loss that no regeneration recipe fully reverses.

7 Causes of Silica Gel Formation and How to Fix Each One

Work through the list in order of likelihood: in field audits, causes 1, 2, and 5 explain most sudden-onset events, causes 3, 4, and 6 most slow-burn ones. Confirm the diagnosis with test work before committing capital.

1. Fast acid addition with local low-pH zones. Symptom: gel and hard scale form on agitators, baffle plates, and pipe walls nearest the acid addition point; gelation worsens at high acid strength. Mechanism: concentrated acid striking the slurry creates microzones that are simultaneously hot, strongly acidic, and highly supersaturated — silicate dissolution is fastest there, and the moment those microzones mix back into the bulk and dilute, the dissolved silica flash-polymerizes. Fix: add acid in stages at diluted strength, ensure vigorous impeller coverage of the addition point, and reconcile the reagent balance against the acid consumption calculation method so addition rate tracks feed mineralogy rather than a fixed setpoint.
2. High leach temperature with uncontrolled cooling. Symptom: leaching itself runs fine, but gel and scale appear in transfer launders, flash lines, and heat exchangers as liquor cools. Mechanism: silica solubility rises with temperature, so hot liquor carries more dissolved silica; on cooling, that liquor becomes supersaturated and the excess either scales hot surfaces or polymerizes in the bulk. Fix: control the cooling profile deliberately, keep liquor above its saturation point until after solid-liquid separation, insulate transfer lines, and avoid thin cold-wall surfaces where scale nucleates fastest.
3. Excessive grind liberating silicates. Symptom: gelation appears or worsens after a mill upgrade, a finer product specification, or a switch to more refractory feed. Mechanism: finer grinding increases both the liberated silicate surface area (faster dissolution kinetics) and the colloidal slime load, so dissolved silica arrives faster than the circuit can hold it below saturation. Fix: re-optimize the target grind against leach kinetics using a shrinking-core model framework — accept a slightly slower leach in exchange for a filterable pulp — and deslime the grind before leaching.
4. High solids density shortening gel induction time. Symptom: gelation appears when operators push throughput by raising pulp density, or after a blend shift increases feed fines. Mechanism: higher solids density releases more silica per liter of liquor and shortens diffusion paths, so the induction period before visible gelation contracts; rising viscosity also accelerates particle collision rates. Fix: cap operating solids density at the value confirmed by bench gelation tests, and dilute rather than starve the filter when rate falls.
5. Insufficient aging time before filtration. Symptom: liquor filters acceptably from the early leach tanks but blinds the filter when fed straight from a short-residence agitated tank; problems appear after residence time is cut. Mechanism: polymerization is still in its fastest phase when the liquor reaches the cloth, so the filter does the de-gelling instead of the process. Fix: pick one policy deliberately — either extend controlled aging so polymerization completes under conditions where colloids can be flocculated, or shorten liquor residence and filter immediately. The worst position is uncontrolled residence in between.
6. Recycling silica-saturated liquor without seeding control. Symptom: problems grow week over week; the circuit behaves worse after barrens or wash-liquor recycle increases. Mechanism: recycle accumulates dissolved silica toward saturation, and an unseeded, near-saturated liquor can then polymerize in a sudden burst when one variable (temperature, pH, solids) tips it over. Fix: monitor dissolved SiO2 on the recycle stream, bleed a controlled purge, and where recycle is essential, maintain a consistent seed population (fine gypsum or aged silica) so polymerization deposits on particles instead of building gel in the bulk.
7. Neutralization stepping through the gelation window. Symptom: gel appears in the neutralization vessel or the first CCD stage rather than in the leach itself. Mechanism: raising pH from strong acidity toward neutral drags the liquor through the pH 2-6 band where polymerization is fastest, while the liquor still carries its full silica load. Fix: where the flowsheet allows, separate solids before bulk neutralization so the liquor neutralizes without a solids bed to foul; otherwise pass through the window quickly at high mixing energy with flocculant ready ahead of the settlers. Detailed neutralization chemistry is treated in a companion article; the leach-side duty is to know where your liquor sits on the pH axis at every step.

Remedies Ranked by Cost and Lead Time: What to Do First

Remedies differ by orders of magnitude in cost, so sequence them. Silica gel formation responds to the first two tiers in most cases, and test work (next section) tells you which tier your ore actually needs. The ranking:

Tier 1 typically recovers 20-50% of lost filtration capacity (indicative) when the root cause sits in operating practice. Tier 2 removes the fastest-dissolving, most colloid-generating fraction before it meets acid; flocculant and filter aid selection — anionic polyacrylamide type and dose for gel-laden solids, plus a compressible precoat or body-feed strategy — protects existing filters rather than replacing them. Tier 3 choices should be confirmed at representative scale, as with our pilot-scale plate-and-frame filter presses. Tier 4 — leach the easily dissolved fraction, separate and wash the liquor before the gel window opens, then advance residues to a second stage — is the definitive structural fix for genuinely high-silica feeds and should be decided from pilot data, not plant troubleshooting.

The cost of doing nothing is worth stating in the same units. Filtration capacity loss of 30-70% of design flux (typical, indicative) converts directly into throughput loss or added filter area; flocculant consumption commonly rises from a baseline of 20-40 g/t toward 60-150 g/t (indicative) as gel-laden solids defeat normal dosing; filter aid consumption rises in proportion to the blinding rate; and each major cloth-cleaning or scale-removal shutdown typically costs 8-48 hours of circuit downtime (indicative). All figures are order-of-magnitude planning numbers to be confirmed against your own plant data.

How Do You Confirm the Cause with Test Work?

Troubleshooting without measurement is guesswork, and the specific driver of silica gel formation is unusually measurable at bench scale. Three levels of test work cover the need:

1. Bottle roll with filtration-time tracking. Run the leach as designed, and every 30-60 minutes pull a timed aliquot through a small vacuum filter or pressure cell, recording filtration time to a fixed volume. A filtration time that doubles between two pulls is the earliest objective alarm — it beats viscosity and turbidity as a leading indicator. Vary one parameter at a time (acid strength, temperature, solids density, aging time) to isolate causes 1 through 5.
2. Viscosity measurement. Track apparent viscosity on the same aliquots with a rotational viscometer at a fixed shear rate. Rising viscosity ahead of the filter confirms bulk polymerization; stable viscosity with falling filterability points instead to colloids and cloth blinding, which changes the remedy entirely.
3. Pilot continuous runs. Before committing to a two-stage flowsheet or a desliming retrofit, run the full circuit continuously — leach, age, separate — at design residence times. Bench batch tests under-predict gelation because they lack the recycle accumulation and residence-time distribution of a real circuit; a continuous lab-scale agitated leaching and solid-liquid separation unit is designed for exactly this duty, and its separation results transfer to plant filter sizing far better than bottle-roll end points.

Report every test with pH, temperature, dissolved SiO2, and filtration time together; a gelation conclusion that cannot be tied to one of those four numbers will not survive contact with the plant.

What Do Field Data Typically Show?

Across published case discussions and our own project files, three patterns repeat often enough to serve as screening benchmarks (all figures typical or indicative, to be confirmed per site):

• Filtration flux: circuits with active gelation typically operate at 30-70% of design filter flux, and recovery of 30-50 percentage points after desliming or acid-staging fixes is common in case reports.
• Liquor silica: dissolved SiO2 in leach liquor from clay-rich and saprolitic feeds typically runs 200-800 mg/L during active leaching (indicative) — far above the 100-140 mg/L amorphous solubility at 25 °C, which is only sustainable while leaching stays hot and strongly acidic.
• Viscosity and moisture: apparent liquor viscosity rising 1.5-3× over baseline and filter cake moisture climbing 4-8 percentage points (indicative) are the two symptoms most consistently reported together, and both reverse when the supersaturation driver is removed.

Treat these as screening magnitudes for framing the business case, not as design values; more test protocols are collected in our resource library. A one-week sampling campaign on your own circuit costs little and replaces all three numbers with data you can defend.

If your plant shows two or more symptoms from Table 1, request a quote for a bench-scale gelation and filtration assessment of your current feed blend.

Frequently Asked Questions

What is silica gel formation?
Silica gel formation is the process by which dissolved monosilicic acid (H4SiO4), released from silicate minerals during acid attack, polymerizes through siloxane condensation into polysilicic acid and finally into a continuous gel network. It is triggered by supersaturation rather than total dissolved silica, is fastest in a characteristic pH window, and shows up as filtration rate collapse, cloth blinding, thickener turbidity, and silica-gypsum scale.

At what pH does silica gel fastest?
Amorphous silica solubility passes through a broad minimum near pH 2-3, so dissolved silica sits closest to supersaturation there, and gelation risk peaks as liquor moves into and through roughly pH 2-6. Liquor held below pH 1-1.5 is generally far more stable. The dangerous moments are neutralization steps and dilution events that drag a silica-rich liquor through the window, not the strongly acidic leach itself.

How much dissolved silica is too much in acid leach liquor?
The absolute number matters less than the margin to saturation. Amorphous silica solubility is roughly 100-140 mg/L as SiO2 at 25 °C and rises with temperature, so a hot leach can carry several hundred mg/L indefinitely. The rule of thumb is to stay below about 1.5× saturation for your liquor temperature and chemistry (indicative). Track dissolved SiO2 routinely and act when it climbs toward saturation.

Does higher leach temperature make silica gelation worse or better?
Both, in different places. Hotter liquor dissolves silica faster and holds more of it in solution, so leaching improves. The risk appears when that silica-rich liquor cools: solubility falls, the liquor becomes supersaturated, and the excess either scales the cooling surfaces or polymerizes in the bulk. High temperature is an ally only if liquor is separated from solids and kept hot, or cooled in a controlled way, before filtration or neutralization.

Can filter aid fix a silica-blinded filter cloth?
Filter aid protects a filter that is handling colloidal silica; it cannot rescue one that is already blinded. A precoat or body-feed builds a porous, sacrificial layer that gel particles smear onto instead of into the cloth, which slows blinding considerably. Once the cloth itself is fouled with bonded silica gel, only chemical cleaning or replacement recovers flux. Treat filter aid as a tier-two mitigation while fixing the supersaturation driver upstream.

Why does gelation appear after neutralization instead of during leaching?
Strongly acidic liquor below roughly pH 1-1.5 is kinetically stable even when supersaturated, because polymerization is slow there. Raising pH during neutralization drags that liquor through the pH 2-6 window where polymerization is fastest, while it still carries its full silica load. The gel therefore forms at the neutralization tank or first CCD stage. Fix it by separating solids before bulk neutralization, or passing through the window quickly with seed and flocculant ready.

How do I test my ore for silica gelation risk before committing to a flowsheet?
Run bottle-roll leaches on representative core at design acid strength and temperature, and track dissolved SiO2, pH, viscosity, and vacuum filtration time on timed aliquots. Confirm mineralogy — clay, serpentine, talc, volcanic glass content — because reactive silica, not head grade, drives risk. Then run a continuous pilot with recycle to capture accumulation effects before finalizing choices such as desliming cuts or two-stage leach.

Conclusion

Silica gel formation is a solubility-control problem, not a filtration-equipment problem: manage dissolved silica away from supersaturation — through blending, desliming, staged acid addition, and controlled aging — and the filters, thickeners, and exchangers will follow. Start this week by adding dissolved SiO2 and filtration-time tracking to your leach sampling program, then confirm the root cause with the bottle-roll protocol above before spending capital.

References

• Iler, R. K. The Chemistry of Silica: Solubility, Polymerization, Colloid and Surface Properties, and Biochemistry. John Wiley & Sons.
• Wills, B. A., and Finch, J. A. Wills’ Mineral Processing Technology, 8th edition. Elsevier.
• Weiss, N. L. (ed.). SME Mineral Processing Handbook. Society for Mining, Metallurgy, and Exploration.
• Free, M. L. Hydrometallurgy: Fundamentals and Applications. John Wiley & Sons.
• Fuerstenau, M. C., and Han, K. N. (eds.). Principles of Mineral Processing. Society for Mining, Metallurgy, and Exploration.
• Crundwell, F. K., Moats, M. S., Ramachandran, V., Robinson, T. G., and Davenport, W. G. Extractive Metallurgy of Nickel, Cobalt and Platinum Group Metals. Elsevier.
• Havlik, T. Hydrometallurgy: Principles and Applications. Smithers Rapra.
• U.S. Geological Survey. Mineral Commodity Summaries (nickel and cobalt sections). Published annually.

Changelog

• 2026-11-02: First publication.