Entrainment in Mixer Settler: Causes & Emulsion Fixes

Last updated: 2026-10-23

Entrainment in mixer settler circuits is one of the most controllable sources of solvent loss: a healthy copper SX plant typically holds organic-in-aqueous entrainment below 50 ppm (mg/L) on a volumetric carry-over basis. When levels climb past 100–200 ppm, solvent inventory bleeds out, product purity degrades, and crud growth accelerates. This guide covers the root causes and the fixes that work in the field, with the full separation overview available on our liquid-liquid separation pillar page.

Key Takeaways

• Entrainment above 100 ppm organic-in-aqueous signals a mixing, solids, or temperature problem.
• Phase disengagement time under 60 s is the practical health check for settler hygiene.
• Cutting mixer power input toward 1.5–2.5 kW/m³ usually recovers solvent within weeks.
• Fine solids and crud, not chemistry alone, stabilise most persistent emulsions.
• Centrifugal extractors fix gravity-separation limits but cannot rescue a badly formulated organic.

What Entrainment and Emulsion Actually Are

Before diagnosing anything, define the failure modes precisely, because they are routinely confused on plant floors. Entrainment is a bulk liquid carry-over mechanism that drags droplets of one phase into the other after the phases have had the opportunity to separate. Organic-in-aqueous entrainment sends droplets of extractant and diluent out with the raffinate; aqueous-in-organic entrainment carries acid, iron, and soluble impurities forward into the strip section. Both directions matter, and they usually respond to different remedies.

Solvent Extraction

how Mixer-Settler Settler Area and Mixer Volume Calculator is decided

Emulsion is a metastable fine dispersion that resists coalescence because droplets are small, the interfacial film is stabilised, or both. In solvent extraction (SX) terms, a moderate dispersion in the mixer is desirable — it creates the interfacial area that mass transfer needs. The problem is a dispersion that refuses to break in the settler. Crud, by contrast, is a stabilised interfacial solids film that accumulates where fine particles, precipitates, and organic meet at the aqueous-organic interface. The name originates from early nuclear-industry operations, and every mature SX operator treats crud management as a permanent housekeeping duty rather than a one-off cleanup.

The physics is straightforward. Droplets rise or settle according to Stokes’ law, so settling velocity scales with the square of droplet diameter and inversely with continuous-phase viscosity. A droplet that is 100 µm settles roughly four times faster than one at 50 µm, which is why an over-agitated mixer can silently double the load on a settler that was sized correctly. This is also why entrainment in mixer settler operation must be read as a system symptom rather than a single-point fault: mixing, solids, and temperature all act on the same droplet population, in the same vessel, at the same time. Interfacial tension governs how easily droplets coalesce; solids that gather at the interface — silica, gypsum, clays, precipitated iron or aluminium compounds — form a mechanical barrier that pins droplets in place, producing the classic rag layer that thickens from a few millimetres to half a metre over weeks. If your hardware itself needs review while you troubleshoot, start with our mixer-settler overview before changing chemistry.

Diagnostic Parameters and Limits That Matter

Troubleshooting starts with measurement, not opinion. Four plant measurements will locate most entrainment problems within a week, and all of them can be logged with basic instrumentation and discipline.

Phase disengagement time (PDT). Take a mixed sample in a graduated cylinder and time the interval until the interface reaches a stable position. A clean, well-balanced system shows PDT below about 60 s; 60–120 s is a warning band; above 120 s the settler is being asked to do a coalescer’s job. Trend PDT daily — it is the earliest sensitive indicator, reacting to temperature, organic condition, and solids load days before entrainment numbers move in the raffinate assay.

Entrainment levels. Organic-in-aqueous is best measured by solvent recovery on a filtered sample (gravimetric or infra-red method, reported in ppm or mg/L — specify which, since they differ for diluted organics). Aqueous-in-organic is tracked via water content (Karl Fischer) or by chloride/sulphate carry-through into the strip liquor. Typical operating targets used across the industry are summarised below; treat them as starting points, not specifications.

Mixer power input. Power per mixer volume is the single most under-instrumented variable in SX. Plants that install a wattmeter or power transducer on mixer drives usually discover they are running 30–50% above the intensity their chemistry actually needs.

Temperature and viscosity. Viscosity of both phases falls sharply with temperature, and coalescence improves with it. A plant that drops from 35 °C to 18 °C overnight in winter can watch PDT triple without any other change. Where geometry and diluent flammability allow, holding the settler at a steady 30–40 °C is cheap insurance; check the diluent flash point first and consult the design basis in our mixer-settler design notes.

Organic condition and sampling discipline. Two supporting measurements complete the diagnostic set. Interfacial tension, measured on submitted organic and aqueous samples, falls as an organic ages or accumulates degradation products; a falling trend with stable mixing points to chemistry rather than hardware. Separately, respect sampling discipline: sample away from settler weirs and outlets where local turbulence exaggerates carry-over, use the same point every day, and filter or not filter consistently. An unfiltered grab sample reads crud fragments as entrainment; an over-filtered one hides it. Neither is wrong, but switching between them makes the trend line meaningless.

Two cautions on limits. First, every number above depends on the extractant, diluent, modifier package, and metal system — a nickel laterite system with a carboxylic acid extractant will not share limits with copper D2EHPA-style circuits. Second, ppm and mg/L are only equivalent when the dispersed phase density equals water’s; diluted organic at ~0.8 g/cm³ means 100 mg/L gravimetrically is ~125 ppm volumetrically. State your basis on every trending chart.

Entrainment in Mixer Settler Equipment: Root Causes and How to Separate Them

Persistent entrainment in mixer settler circuits almost always traces to one or more of six causes, and the diagnosis is a pattern-matching exercise: each cause leaves a distinct symptom fingerprint in the settler. The table below is the working checklist our engineers take onto site.

The mixing-intensity case deserves emphasis because it is the most common and the cheapest to fix. Every impeller has a trade-off: droplet shear sets interfacial area for extraction kinetics, but excessive tip speed shatters droplets below the size the settler can handle. Many plants inherited aggressive impeller settings from commissioning and never revisited them once chemistry stabilised. Poor control of the organic-to-aqueous (O/A) ratio interacts with this badly — when the ratio drifts, a mixer designed for organic-continuous operation can flip to aqueous-continuous locally, and the droplet size distribution collapses. Our O/A ratio in SX note covers the control scheme side in detail.

Fine solids deserve their own warning. In hydrometallurgical circuits, solids arrive with pregnant leach solution (PLS), form in situ as gypsum or basic ferric compounds, or enter as dust and clay through open tanks. Once solids reach the interface they stabilise crud, and crud then acts as an emulsion factory: it pins droplets, traps organic, and grows whenever solids loading spikes. Managing it is a dedicated discipline — see our crud treatment in SX resource for removal and treatment options. Finally, note that extractant degradation compounds the picture: an aged organic (for example, hydrolysed di-2-ethylhexyl phosphoric acid, D2EHPA, in impurity-removal duty) loses interfacial tension headroom, so the same mixing energy that was benign when the organic was fresh now produces emulsion. If you are weighing extractant choices while troubleshooting, the D2EHPA vs Cyanex 272 comparison is a useful reference for cobalt/nickel duties.

Which Fixes Remove Entrainment in Mixer Settler Circuits — and in What Order?

Work through the fixes in order of cost, not sophistication. The first three actions on this list have solved more entrainment problems across the industry than any capital project.

1. Trim mixer RPM and tip speed. Symptom: cloudy raffinate and PDT creeping upward. Cause: droplets sheared below settling size. Remedy: reduce impeller speed in 5–10% steps, re-running the PDT cylinder test each time, and stop at the speed where extraction efficiency is preserved but PDT recovers.
2. Correct the O/A ratio and phase continuity. Symptom: erratic interfaces and stage-to-stage differences. Cause: ratio drift flipping continuity. Remedy: recalibrate flow instruments, re-establish organic-continuous operation in the mixer, and alarm on ratio deviation beyond ±5%.
3. Remove fine solids before the settler. Symptom: weekly rag-layer growth. Cause: solids arriving in PLS or formed in-circuit. Remedy: polish the feed (sand filters or clarification), pin the interface level lower so crud is skimmed deliberately, and wet-hay the settler walls on a schedule.
4. Skim and treat crud on schedule. Symptom: stable dark interface band with organic hold-up. Cause: mature crud acting as an emulsion stabiliser. Remedy: physical removal, then wash or centrifuge the crud to recover trapped organic before disposal.
5. Raise and hold temperature. Symptom: seasonal or diurnal PDT swings. Cause: viscosity rise at low temperature. Remedy: target a steady 30–40 °C within diluent flash-point limits, using heat balance from the leach circuit where possible.
6. Install coalescence internals in the settler. Symptom: PDT acceptable in tests but entrainment persists at scale. Cause: short effective settling path or surface carry-over. Remedy: baffles, picket fences, and lamella packs that calm flow and promote coalescence.
7. Rebuild the organic inventory. Symptom: chronic entrainment history with degraded organic. Cause: accumulated oxidation products and dissolved impurities. Remedy: bleed and replace a fraction of the organic, add clay treatment or acid wash where indicated, and re-baseline.

Whatever the fix, close the loop with the same instruments that found the problem: PDT daily, entrainment assays weekly, and power input continuously. Fixes applied without a re-baseline cannot be defended in a budget review, and partial fixes that look successful for a month often relapse once solids loading or temperature drifts back.

A note from field experience only: the modest RPM cut in item 1 is routinely the highest-return action in the entire sequence. On a typical plant, trimming mixer power by 20–30% recovers solvent into the circuit within weeks — the entrained organic leaving in the raffinate is inventory you already bought, and every litre that returns is a litre of fresh make-up you do not purchase. Only once mixing is right should capital options be considered.

Centrifugal extractor emulsion: when spinning helps and when it does not

A centrifugal extractor replaces gravity with a centrifugal field of several hundred to a few thousand g, cutting contactor residence time to seconds and shrinking footprint dramatically. It is the correct choice where the chemistry is clean, the phases disengage quickly under force, and space or inventory hold-up is expensive. But centrifugal extractor emulsion behaviour has a hard limit: the machine separates phases, it does not destabilise an emulsion. If solids-stabilised crud or a degraded organic is pinning droplets, a centrifuge simply concentrates the rag layer against its weir faster, and you trade settler problems for frequent bowl cleaning. Pilot both options on your real liquor before committing — the escalation path, including equipment selection guidance, is in our centrifugal extractor resource, and our pilot testing service exists precisely to settle this choice with data.

How Much Do Solvent Losses to Entrainment Cost?

The arithmetic is uncomfortable, which is why it is worth doing in writing. Take an illustrative copper SX train with 100 m³/h of raffinate leaving the extraction settler carrying 100 ppm (volumetric, ~80 mg/L gravimetric basis given a 0.80 g/cm³ organic) of entrained organic. That is 10 L/h, or 240 L/day, or roughly 87 m³ of organic per operating year at 8,400 h. At an illustrative delivered organic cost of 1.50 USD/L, the loss is about 131,000 USD per year — before counting disposal costs for oil-bearing raffinate, downstream electrowinning (EW) cathode risk from dissolved organics in copper SX-EW plants, and the safety exposure of oil on pond surfaces. Scale the estimate linearly with flow and entrainment level; a 500 m³/h plant at 200 ppm faces a seven-figure annual exposure.

The table deliberately uses conservative organic pricing and excludes hidden costs: crud disposal, increased fire risk, and cathode rejections from organic burn-in during EW. In audits we have reviewed, the hidden costs frequently match the visible solvent invoice. The conclusion is not subtle — entrainment is a profit leak with a straightforward plumbing fix, and the payback on a single set of mixer-speed and interface-level corrections is usually measured in months, not years.

What Does a Successful Entrainment Fix Look Like in Practice?

The following composite case is illustrative, assembled from typical patterns rather than any single named client. A copper SX circuit running 150 m³/h of PLS through conventional mixer-settlers reported persistent organic loss: raffinate assays showed organic-in-aqueous entrainment of 180 ppm against a 50 ppm target, and phase disengagement time had drifted from 60 s to 140 s over two years.

Diagnosis followed the sequence above. Interface solids assays showed a rag layer dominated by fine gypsum and silica; temperature logs showed winter operation at 17 °C against a design 35 °C; and a power survey found mixers drawing 3.4 kW/m³ against a typical 2.0 kW/m³ design intensity. The response was deliberately unheroic: impeller speed cut 18%, temperature raised and held at 32 °C, O/A ratio instrumentation recalibrated after a ±15% drift was found, and a monthly crud skimming schedule established.

Within six weeks, illustrative results were: entrainment down from 180 ppm to 40 ppm; PDT down from 140 s to 55 s; and organic make-up purchases down roughly 35% against the same quarter the previous year. No capital equipment was purchased. The entire intervention was measurement, adjustment, and housekeeping — which is exactly why it started in week one rather than waiting for a budget cycle. The lesson generalises: entrainment in mixer settlers is rarely a single-cause failure, and the plants that recover fastest are the ones that fix the cheap variables first and verify each step with numbers before moving to the next.

Frequently Asked Questions

What entrainment level is acceptable in copper SX?
Most copper SX operators work to below 50 ppm organic-in-aqueous in raffinate as a healthy target, with 50–100 ppm indicating that mixing, temperature, or solids need attention. Above 200 ppm, solvent losses become financially material and downstream electrowinning quality is at risk. Aqueous-in-organic is typically held below a few hundred ppm to limit acid and impurity carry-through into the strip circuit. Targets should be set against your own extractant and diluent system.

What is the difference between emulsion and crud?
Emulsion is a dispersion of fine droplets of one phase suspended in the other that resists coalescence; it is a bulk-liquid phenomenon driven by mixing energy, interfacial tension, and viscosity. Crud is a solids-stabilised film that forms at the liquid-liquid interface, composed of fine particles bound with organic. Crud acts as an emulsion stabiliser: it pins droplets at the interface, thickens over time, and traps valuable organic that must be recovered during removal.

Does lowering mixer RPM always reduce entrainment?
Lowering impeller speed reduces droplet shear and usually improves phase disengagement, but it is not unconditional. Extraction kinetics need interfacial area, so cutting speed too far sacrifices mass transfer and stage efficiency. The correct method is a stepped trial: reduce speed in 5–10% increments, re-testing phase disengagement time and extraction efficiency at each step, and settle at the lowest speed that preserves mass transfer. On most plants the optimum sits well below commissioning settings.

How do I measure organic-in-aqueous entrainment?
The common methods are solvent recovery on a filtered, acidified raffinate sample, reported gravimetrically (mg/L) or by infra-red absorbance calibrated against your own organic. Specify the basis on every result: volumetric ppm, mg/L, and g/m³ differ, particularly for a diluted organic at ~0.8 g/cm³. Sample at a consistent point away from settler weirs, filter or not consistently, and trend daily — a stable method matters more than an absolute-accurate one for troubleshooting.

When should I choose a centrifugal extractor over a mixer-settler?
Choose a centrifugal extractor when phases disengage well under applied force but gravity settling is too slow for the flow, when footprint or solvent inventory must be minimised, or when short contact time protects a sensitive reagent. Avoid it as an emulsion cure: a centrifuge separates phases but cannot destabilise a solids-stabilised emulsion, so a crud-driven problem will simply migrate into bowl cleaning. Always pilot-test on your actual liquor before selecting between the two technologies.

Can entrained organic damage cathode quality in EW?
Yes. Organic entrained into loaded electrolyte decomposes at the cathode surface during electrowinning, producing burn spots, voids, and warping that downgrade cathode to lower purity grades. This is why serious circuits include floatation or coalescing media polishing between extraction and stripping, and why aqueous-in-organic limits matter even though the organic itself is your own inventory. A cathode quality complaint is often the first visible symptom of an entrainment problem that started weeks earlier upstream.

Why does entrainment get worse in winter?
Lower temperature raises the viscosity of both aqueous and organic phases and slows droplet coalescence, so phase disengagement time lengthens and entrainment rises even with mixing unchanged. The effect is steep: a plant dropping from 35 °C to 18 °C can see disengagement times triple. Where diluent flash point and process heat balance allow, holding settlers at a steady 30–40 °C is one of the cheapest entrainment controls available, and it also stabilises extraction equilibrium across the day.

Conclusion: Entrainment Is a Measurement Problem Before It Is a Capital Problem

Entrainment in mixer settler circuits responds to discipline faster than to capital: log phase disengagement time daily, hold mixer power input near 1.5–2.5 kW/m³, control the O/A ratio within ±5%, keep temperature steady, and remove crud on schedule, and most plants recover their solvent inventory within weeks. Only after those levers are exhausted should you weigh coalescence internals or a centrifugal extractor — and that choice deserves pilot data, not a brochure. Start with our pilot testing service to validate fixes on your own liquor, or browse the rest of our technical resources for design guides on every stage of the circuit. The solvent you stop losing this quarter pays for the program.

References

1. Rydberg, J., Cox, M., Musikas, C., & Choppin, G. R. (eds.) (2004). Solvent Extraction Principles and Practice, 2nd ed. Marcel Dekker, New York.
2. Lo, T. C., Baird, M. H. I., & Hanson, C. (eds.) (1983). Handbook of Solvent Extraction. John Wiley & Sons, New York.
3. Ritcey, G. M., & Ashbrook, A. W. (1979). Solvent Extraction: Principles and Applications to Process Metallurgy, Parts 1–2. Elsevier, Amsterdam.
4. Gupta, C. K., & Mukherjee, T. K. (1990). Hydrometallurgy in Extraction Processes, Vols. 1–2. CRC Press, Boca Raton.
5. Green, D. W., & Perry, R. H. (eds.) (2008). Perry’s Chemical Engineers’ Handbook, 8th ed., Section 15 (Liquid-Liquid Extraction). McGraw-Hill, New York.
6. Habashi, F. (1993). A Textbook of Hydrometallurgy. Métallurgie Extractive Québec, Québec City.
7. U.S. Geological Survey (2024). Mineral Commodity Summaries 2024 — Copper. U.S. Department of the Interior, Reston, Virginia.

Changelog

2026-10-23 — First published.