Manganese in Copper Electrowinning: 7 Causes and Fixes

Last updated: 2026-11-04

A violet tint in the electrolyte overflow is the classic late symptom of manganese in copper electrowinning: by the time permanganate shows its color, anode fouling has typically already cost 1-3 points of current efficiency. Manganese reaches the tankhouse through the solvent extraction circuit as entrained aqueous and co-extracted trace, then accumulates in the closed electrolyte loop. Seven causes, fixes, and a worked bleed calculation follow.

Key Takeaways

• Manganese reaches the tankhouse through the SX circuit: entrained aqueous, co-extraction traces, and crud carryover.
• At Pb-based anodes, Mn(II) oxidizes to MnO2 scale and violet permanganate, raising voltage and cutting current efficiency.
• The electrolyte loop is closed, so manganese accumulates until you bleed it on purpose.
• Fix sources at the SX interface first; size the electrolyte bleed second.
• Monitor the tint, weekly Mn assays, and current efficiency per cathode harvest.

Manganese in Copper Electrowinning: The Anode Chemistry

Electrowinning is an electrochemical process that plates copper from acidic sulfate electrolyte onto stainless-steel cathode blanks while oxygen evolves at insoluble anodes, and every side reaction competing with oxygen evolution at the anode carries an energy and purity penalty. That is exactly where manganese hurts a copper tankhouse.

At a lead-based anode the useful reaction is water oxidation to oxygen. The problem is that manganese offers the anode easier things to do. Mn(II) oxidizes to manganese dioxide, MnO2, at a standard potential near +1.23 V versus SHE — essentially the same as oxygen evolution — and further to permanganate, MnO4-, near +1.51 V. Because oxygen evolution on Pb-Ca-Sn alloys is kinetically sluggish, real anode potentials in copper EW sit around 1.7-1.9 V versus SHE (indicative), comfortably high enough to drive both manganese products. The consequences stack up in three ways:

• MnO2 forms as a hard adherent scale that buries the active anode surface, raises oxygen overpotential, and pushes cell voltage up.
• Permanganate is a strong oxidant that colors the electrolyte violet and shuttles to the cathode, where it is reduced back to lower manganese states — a parasitic redox cycle parallel to the familiar Fe(III)/Fe(II) shuttle, and each shuttle pass wastes current that should have plated copper.
• Scale that loses adhesion sloughs off as dense black-brown sludge, which piles up in cells and can be kicked onto cathode faces as spotting and nodulation.

Additives suffer too. Guar is a natural polymer additive that improves copper deposit morphology, and it is chemically vulnerable: permanganate and MnO2 surfaces oxidize its backbone, so plants with active manganese see guar consumption drift upward even at constant viscosity control. The interactions spread the damage. Iron compounds the picture because Fe(III)/Fe(II) cycling and Mn redox cycling both burn current, so a plant that only tracks iron will under-explain its current efficiency loss in copper electrowinning. Colloidal silica, common in leach liquors, stabilizes fine MnO2 particles and makes them behave like crud rather than settle out.

The reason this becomes chronic rather than episodic is the flowsheet: tankhouse electrolyte is a closed loop shared with the strip stage. Copper is plated out and water is electrolyzed, but manganese is not removed by electrowinning. Only three exits exist — deliberate electrolyte bleed, MnO2 sludge carted out with cell cleaning, and scale stripped off during anode maintenance. If inputs exceed those exits, concentration rises until something visible breaks.

Where Does Manganese in Copper Electrowinning Electrolyte Come From?

Manganese enters with the ore and reaches the tankhouse through paths that the SX circuit either cannot stop or does not stop by design.

On the ore side. Oxidized copper deposits frequently carry manganese oxides — pyrolusite and cryptomelane among them — as gangue, and manganese carbonates such as rhodochrosite appear in some belts. Under hot acidic leaching, manganese carbonates dissolve directly to Mn(II); manganese dioxide needs a reductant, and the leach supplies them naturally: Fe(II) in solution, sulfide minerals, and organic matter all reduce Mn(IV) to soluble Mn(II) at leach pH. Where concentrates are roasted, manganiferous feed converts to acid-soluble manganese phases, so roast and preg liquors carry Mn(II) forward as well. The result is PLS with manganese anywhere from tens to a couple of thousand mg/L (indicative — assay your own, because the band depends entirely on ore body and blend).

Through the SX circuit. Aldoxime and ketoxime extractants are selective for copper, and true chemical co-extraction of Mn(II) is small. What passes into copper solvent extraction electrolyte is physical and second-order: aqueous droplets entrained in loaded organic, trace co-extraction at high loading, and crud solids. The strip stage then delivers all three into the stripped electrolyte, which is why manganese is best understood as an SX-strip interface problem that the tankhouse pays for.

In the closed loop. Once in the electrolyte, Mn(II) has nowhere to go except the anode or the bleed line. Copper plating does not purge it, acid makeup does not dilute it away at steady state, and the loop keeps concentrating it. Plants commonly run 50-500 mg/L for years without noticing anything beyond slightly dirtier anodes; trouble typically becomes visible above several hundred mg/L (indicative), and the permanganate tint is a late-stage flag, not an early warning.

Treat the table as bands to anchor sampling, not specifications. Two engineering judgments matter more than any single number. First, the electrolyte band is a steady-state outcome, not an input: it is set by ingress versus bleed, so a rising trend tells you the balance has shifted even before any absolute limit is hit. Second, the sludge row is diagnostic gold — if cell muck assays as mostly MnO2, your manganese problem is already an anode problem, and the cost side of this article is no longer hypothetical.

The 7 Causes of Manganese in Copper Electrowinning — and the Fix for Each

Each item below follows the same discipline: the phenomenon you observe, the root cause behind it, and the fix that actually addresses it.

1. Manganese-bearing feed ore or roast liquor. You see PLS manganese step up after an ore-blend change or a new pit bench. The cause is dissolution of Mn oxides and carbonates in the leach, or soluble manganese from roasted manganiferous feed, passing forward in PLS. The fix is upstream: characterize Mn per ore source, blend high-Mn feed down, and manage leach redox so you know what is coming — SX is not designed to stop Mn(II).

2. Aqueous entrainment from the strip stage. You see electrolyte manganese creep in proportion to organic carryover and slow phase disengagement. The cause is PLS and raffinate droplets entrained in loaded organic and released in the strip stage. The fix is entrainment discipline — settling area, mixing energy, wash-stage operation — per the checklist in entrainment in mixer settlers; degraded organic makes disengagement worse, so also review extractant degradation.

3. Crud carryover as a manganese carrier. You see manganese spikes that track visible solids in the organic and at the interface. Crud is a stabilized emulsion of fine solids and organic that scavenges Mn-bearing gangue and travels with the organic phase. The fix is aggressive skimming and interface management, following crud treatment in solvent extraction practice, plus treatment of the crud source rather than repeated manual cleanouts.

4. Excessive anode overpotential. You see violet electrolyte and rising cell voltage together. The cause is anode potential high enough to push Mn(II) beyond MnO2 to Mn(VII), and to drive MnO2 growth on the anode face, which raises potential further — a self-accelerating loop. The fix is electrical and mechanical: avoid current-density spikes, keep anodes flat, straight, and clean, and confirm electrolyte temperature and acidity are at design, both of which shift the potential balance.

5. Insufficient electrolyte bleed. You see slow, relentless accumulation even though the SX interface looks clean. The cause is arithmetic, not chemistry: a closed loop with manganese input and no exit. The fix is a deliberately sized bleed, calculated from measured ingress and a target steady state — the worked calculation is in the next section.

6. Fe-Mn co-chemistry misdiagnosis. You see current efficiency loss blamed entirely on iron, Fe-control fixes applied, and no recovery. The cause is that Fe(III)/Fe(II) and Mn redox cycling produce the same symptom — current consumed without copper plated — so treatment aimed at iron alone leaves the manganese cycle running. The fix is to assay both metals, apportion the loss using the voltage and color signature, and only then spend money.

7. Anode metallurgy and coating choice. You see two comparable tankhouses holding different manganese steady states at the same electrolyte assay. The cause is anode surface metallurgy: hard-rolled Pb-Ca-Sn versus coated titanium DSA-type anodes (stated generically, no endorsement implied) differ in oxygen overpotential and in how tenaciously MnO2 adheres. The fix is a generic anode review — coated titanium generally sheds MnO2 as manageable sludge rather than adherent scale — but this is a whole-of-life economics decision, not a manganese fix on its own.

How Do You Diagnose Manganese in the Tankhouse and Size the Bleed?

Diagnosis works best as a fixed routine, because manganese in copper electrowinning develops slowly and the evidence is easy to normalize away. The protocol below assumes a plant with routine laboratory support; the monitoring anchors are summarized in the table.

1. Establish the baseline. Sample electrolyte at the overflow, jar-test for tint, and send a filtered sample for Mn, Fe, and acid assay. Record current efficiency from the last two harvests and cell voltage by section. You cannot diagnose a trend you have not measured.
2. Mass-balance the inputs. Assay PLS, raffinate, stripped electrolyte, and electrolyte the same week. Calculate manganese ingress in kg/h from the strip-stage balance — this single number drives the bleed calculation in step 7.
3. Localize the transfer path. Distinguish entrainment, co-extraction, and crud: sample loaded organic after settling, test wash-stage efficiency, and inspect the interface for crud. Entrainment-dominated ingress fixes are cheap; crud-dominated ingress means an upstream solids problem.
4. Inspect the tankhouse side. Pull sample anodes and measure scale thickness and weight change; tally sludge volume at recent cell cleans; recheck current efficiency per harvest. This quantifies how much of the problem has already become capital damage.
5. Check additive consumption and the iron interaction. Compare guar dose against baseline at constant viscosity target, and confirm Fe(II)/Fe(III) assays so manganese is not silently credited or debited for iron’s losses.
6. Fix the sources first. Apply the SX-side fixes from causes 1-3 and 6 — ore blending, entrainment control, crud removal, misdiagnosis correction. A bleed sized against an uncontrolled ingress just moves more electrolyte per hour for the same problem.
7. Size the bleed and keep monitoring. Worked example: take a 150 m3/h electrolyte loop with measured ingress of 0.6 kg/h Mn — say 0.3 m3/h of PLS at 2.0 g/L Mn entrained through the strip stage. With no bleed, concentration rises 0.6 kg/h divided by 150 m3/h = 0.004 kg/m3 per hour, about 96 mg/L per day — roughly 1 g/L in ten days. To hold a steady state at a target of 800 mg/L (0.8 kg/m3), the required bleed is B = ingress divided by target = 0.6 / 0.8 = 0.75 m3/h, or 0.5% of circulation. With about 2000 m3 of loop inventory, the time constant is 2000 / 0.75 ≈ 2670 h (about 111 days) to close 63% of the gap, so start the bleed early and verify against weekly assays. Assumptions: well-mixed loop, constant ingress, and no other manganese sinks — deliberately conservative, because MnO2 sludging and anode scale are additional, uncredited removal paths.

What Does Manganese Cost You If You Ignore It?

The bill arrives in four currencies: energy, additives, anodes, and cathode quality. The ranges below are indicative and assume a modern copper EW tankhouse at 280-300 A/m2, 1.9-2.1 V cell voltage, 90-93% current efficiency (typical), with iron losses apportioned separately — state your own basis before quoting any of this upstream.

Current efficiency is the headline. Uncontrolled manganese typically costs 1-3 points of CE (indicative) through the permanganate cathode shuttle, MnO2-driven overpotential, and additive degradation working together. At a 2.0 V baseline, each CE point is roughly 1-2% more energy per tonne of copper, and the voltage rise from MnO2 scale adds another 0.05-0.15 V — on a 2.0 V cell that is 2-8% more kWh/t. These multiply: a plant sliding from 92% to 90% CE while voltage climbs 0.1 V can see specific energy rise 4-8% before anyone changes a setpoint.

Additive consumption drifts quietly. Guar demand up 10-30% (indicative) at constant viscosity control is common where permanganate and MnO2 surfaces attack the polymer, and the same oxidizing environment is hard on any organic that cycles back to the tankhouse. Anode life shortens: heavy MnO2 scaling accelerates lead corrosion mechanisms, and tankhouses with chronic manganese commonly report 20-40% shorter anode replacement cycles (indicative) plus two to four times the cell-cleaning labor on MnO2 sludge. Cathode quality is the visible tail: manganese-driven spotting and nodulation push rejections and re-melting costs that never appear in the power bill.

The comparison verdict: a sized bleed and SX-side fixes cost a rounding error against the energy and anode line alone, and every month of delay adds inventory that takes a hundred days or more to bleed out. Ignoring manganese is not free — it is deferred spending at a markup.

Worked Plant Example: Tracing a Current-Efficiency Slide to Manganese

A 55,000 t/a copper SX-EW operation changed its ore blend toward a manganiferous bench and, over four months, watched tankhouse electrolyte manganese climb from 120 to 780 mg/L while current efficiency fell from 92.6% to 90.4% — a 2.2-point slide worth roughly 5% more energy per tonne of cathode. Cell voltage rose 0.09 V at constant current, guar dose drifted up 16% at constant viscosity control, and cell cleaning went from once to three times in the period, with muck assaying as MnO2-dominant. Sample anodes carried up to 2 mm of hard scale.

The diagnostic followed the 7-step procedure. The mass balance showed 0.6 kg/h of manganese entering through the strip stage — 0.3 m3/h of entrained PLS at 2.0 g/L Mn — and the wash-stage efficiency test localized it to entrainment after a mixer impeller speed-up that had pushed the dispersion band up. Co-extraction and crud were minor contributors.

Fixes were sequenced: impeller speed and layout corrected, wash stage re-tuned, crud skimmed and its upstream solids source treated, and a 0.75 m3/h bleed installed per the calculation in the diagnostic section to hold roughly 650-800 mg/L. Within two harvests, current efficiency recovered to 92.4%, cell voltage returned to baseline, guar dose settled back within 5% of its historical value, and the permanganate tint disappeared from the overflow. For context on where this duty sits, see our copper industry applications overview.

The lesson generalizes: the tankhouse symptom was real, but the root cause lived at the SX-strip interface, and the cheapest fix was flow discipline, not chemical treatment.

FAQ

Is manganese always harmful in copper EW electrolyte?

Not at low levels. Plants commonly operate for years at tens of mg/L (indicative) with no measurable effect on anodes or current efficiency. Harm scales with concentration and time: the mechanisms — MnO2 scale, permanganate shuttle, additive attack — are all dose-driven. Watch the trend rather than the absolute value, set an alert level near 500 mg/L, and act on ingress before the tint or voltage signature appears.

What does the violet color in the electrolyte overflow actually mean?

It is permanganate, Mn(VII), formed at the anode from Mn(II). Its appearance means anode potentials are high enough to push manganese past MnO2 to MnO4-, which usually coincides with existing MnO2 scale and rising overpotential. Treat it as a late-stage flag: jar-test each shift, confirm with a filtered Mn assay, and work the diagnostic procedure rather than dosing anything to decolorize the symptom.

How fast does manganese accumulate in a closed electrolyte loop?

Fast enough to matter. In the worked example, 0.6 kg/h of ingress into a 150 m3/h loop raised concentration about 96 mg/L per day — roughly 1 g/L in ten days with no bleed. Even small persistent inputs accumulate steadily because electrowinning removes copper, not manganese. Bleed-out is the slow side: with a 0.75 m3/h bleed on 2000 m3 of inventory, the time constant is about 111 days.

Can the existing iron bleed control manganese too?

Usually yes, if it is sized for it. The bleed line does not care which impurity it removes; the question is arithmetic — bleed rate times electrolyte concentration must equal total impurity ingress. Run the calculation with combined Fe and Mn ingress and confirm the existing bleed is large enough. If holding both targets overloads upstream reagent circuits, add a dedicated slipstream rather than compromising iron control.

Does manganese harm stainless-steel cathodes or cathode copper quality?

Indirectly, yes. MnO2 sludge and fines kicked up by gas evolution can lodge on cathode faces as spotting and nodulation, pushing rejection risk. The permanganate shuttle and degraded guar also worsen deposit morphology. The stainless blanks themselves are not corroded by manganese at EW conditions, but keep them clean during harvesting and watch physical copper quality as part of the manganese monitoring protocol.

How do I tell manganese losses apart from iron losses in current efficiency?

Assay both metals and read the signature together. Iron alone burns current through the Fe(III)/Fe(II) shuttle without color change; active manganese adds violet tint, MnO2 sludge, rising voltage, and guar drift. A CE loss with clean color and normal voltage points at iron; the same loss with tint and voltage rise points at manganese. Apportion by fixing one variable at a time and tracking the next two harvests.

Do coated titanium anodes solve the manganese problem?

They change it, not eliminate it. Coated titanium DSA-type anodes (stated generically) run lower oxygen overpotential and shed MnO2 as removable sludge instead of adherent scale, which reduces voltage rise and anode damage. But they do nothing about ingress — the manganese still crosses the strip stage and still accumulates without a bleed. Treat anode choice as one lever inside the full program, evaluated on whole-of-life economics.

Conclusion

Manganese in copper electrowinning is a predictable consequence of a closed electrolyte loop fed by an open SX interface: whatever manganese crosses the strip stage in entrained aqueous, co-extracted traces, or crud stays until you remove it, and the anode turns it into scale, sludge, permanganate, and lost current efficiency. The control logic is always the same order — measure ingress, fix the sources at the SX boundary, then size the bleed to the steady state you want, and monitor with a fixed protocol so the trend never outruns you again. If your tankhouse is showing the tint, the voltage drift, or the guar creep described here, request a quote for a manganese diagnostic review, or start with our copper sector hub to see the full equipment and process support we offer around this problem.

References

• Davenport, W.G., King, M.J., Schlesinger, M.E. and Biswas, A.K., Extractive Metallurgy of Copper, 5th ed., Elsevier.
• Kordosky, G.A., “Copper Recovery Using Leach/Solvent Extraction/Electrowinning Technology: Forty Years of Innovation,” JOM.
• Free, M.L., Hydrometallurgy: Fundamentals and Applications, John Wiley & Sons.
• Sole, K.C., Feather, A.M. and Cole, P.M., published work on impurity transfer and control in copper solvent extraction circuits, Hydrometallurgy and related conference proceedings.
• Guerra, E. and Dreisinger, D.B., study of factors affecting current efficiency in copper electrowinning, published journal article.
• Bruckard, W.J., Senanayake, G. and colleagues, studies on manganese chemistry and removal in hydrometallurgical process liquors, Hydrometallurgy.
• Robinson, T.G. and colleagues, operating surveys and plant data for copper electrowinning tankhouses, published industry conference proceedings.

Changelog

• 2026-11-04: Initial publication.

Related equipment

• Lab-Scale Copper Electrowinning Cell — Electrolysis & Electrowinning