TL;DR #
A three-step wet-chemical purification process — manganese powder displacement, sodium sulfide precipitation, then sodium fluoride polishing — reduces copper from 20.43 g/L to below 5 mg/L and delivers a final manganese sulfate solution at ~102 g/L that meets battery-grade purity specifications. For buyers sourcing NMC precursor materials or evaluating closed-loop recycling suppliers, this defines the minimum process competency you should be demanding from any vendor claiming battery-grade output. Before your next RFQ, ask suppliers to demonstrate specific residual metal concentrations at each process stage — not just a final CoA.
Overview #
Battery-grade manganese sulfate is one of those inputs where the purity gap between “close enough” and “actually qualified” has a direct impact on cathode electrochemical performance, and most procurement teams find this out after a failed incoming QC lot rather than before. The research reviewed here was conducted at an industrial-scale process development facility working with real ternary lithium battery recycling streams — not synthetic lab solutions — using a 20–30 L titanium reactor setup with controlled heating, mechanical stirring, and multi-point sampling across all three purification stages. The starting material was genuinely difficult: manganese at ~97 g/L, copper at 20.43 g/L, cobalt at 647.73 mg/L, zinc at 127.78 mg/L, and calcium at 171.42 mg/L — concentrations that would immediately disqualify the raw solution for any downstream cathode application.
This work is directly relevant to buyers evaluating Cell Selection & Sourcing from Chinese manufacturers who are increasingly integrating recycled precursor streams into their supply chains. Understanding the purification process behind battery-grade manganese sulfate gives you the technical vocabulary to separate credible claims from marketing language.
Three-Step Purification of NMC Recycling Waste Streams: Process Parameters and Performance #
The process sequence matters here. You cannot jump straight to sulfide precipitation on a solution carrying 20 g/L of copper — the reagent cost alone would be prohibitive and you’d lose significant manganese to co-precipitation. The logical order is displacement first, then deep sulfide polishing, then calcium/magnesium removal.
Stage 1: Manganese Powder Displacement for Copper Removal
Metallic manganese displaces copper via a straightforward cementation reaction (Mn + CuSO₄ → MnSO₄ + Cu↓), which is elegant because it adds manganese to the solution rather than a foreign metal ion. The key process variables are reagent stoichiometry and reaction time.
At exactly 1.0× the theoretical manganese requirement, copper drops below 10 mg/L. Adding more than 1.0× theoretical does further reduce copper, but the marginal benefit disappears quickly and excess manganese represents a direct yield loss. Reaction time follows a clean kinetic profile: from 0.5 h to 2 h, copper concentration falls from 134.6 mg/L to 3.84 mg/L. Beyond 2 h, the curve flattens — reaction equilibrium is reached. Optimal parameters: 1.0× theoretical Mn powder dose, 2 h reaction time, ambient temperature, with pH rising from 2.0–2.5 to 4.5–5.0 as the reaction proceeds.
The copper sludge recovered at this stage is not waste. After drying, it contains approximately 50% copper by mass and 10% manganese. A simple dilute sulfuric acid wash removes the manganese fraction, pushing copper content above 70% — recoverable as a commercial byproduct. Copper recovery rate exceeds 70% by this route.
Stage 2: Sodium Sulfide Deep Removal of Residual Heavy Metals
After Stage 1 filtration, the solution still carries residual copper at ~9.43 mg/L, cobalt at 496.6 mg/L, zinc at 25.98 mg/L, nickel at 2.71 mg/L, and cadmium at 3.5 mg/L. Sodium sulfide precipitates all of these as insoluble metal sulfides (CoS, ZnS, CuS, NiS), which is well-established chemistry — but the process window is narrower than it looks on paper.
At 8 g/L Na₂S dose and 30 min reaction time at ambient temperature (~20°C), residual concentrations reach: Cu²⁺ = 2.62 mg/L, Co²⁺ = 1.85 mg/L, Zn²⁺ = 1.28 mg/L, Ni²⁺ = 1.71 mg/L, Cd²⁺ = 0.75 mg/L. All below 3 mg/L — the target threshold for battery-grade feed.
Here is where the process gets counterintuitive, and where I’ve seen suppliers make a costly mistake: extending reaction time beyond 30 minutes causes re-dissolution (返溶) of cobalt and zinc. At 4 h reaction time, residual concentrations of Co and Zn climb back measurably. The dissolution mechanism involves excess S²⁻ forming soluble thio-complexes. This is not a theoretical concern — the data shows the effect clearly at 1 h, 2 h, and 4 h sampling points.
Temperature also matters in a non-obvious way. Higher temperatures drive residual heavy metals lower initially — at 20°C all targets stay below 3 mg/L — but elevated temperature also accelerates re-dissolution. Room temperature is the correct operating point, not elevated temperature.
The Na₂S dosage has a hard upper limit: excess sulfide will begin precipitating Mn²⁺ as MnS, which represents a direct yield loss from your manganese product. 8 g/L is the minimum effective dose and should not be substantially exceeded.
Stage 3: Sodium Fluoride Selective Removal of Calcium and Magnesium
The calcium and magnesium removal step operates through selective precipitation as CaF₂ (Ksp = 2.7 × 10⁻¹¹) and MgF₂ (Ksp = 6.4 × 10⁻⁹). The extremely low solubility products make NaF the correct reagent, but dosage and thermal conditions need careful control — too much fluoride introduces free F⁻ that attacks process equipment and contaminates the final product.
Starting concentrations heading into Stage 3: Ca²⁺ = 171.42 mg/L, Mg²⁺ = 62.90 mg/L, Mn = 102.14 g/L. At optimized conditions (80°C, 4 g/L F⁻ as NaF, pH 5.0, 600 rpm agitation, 1 h reaction, 12 h aging), the final concentrations reach Ca²⁺ = 8.54 mg/L, Mg²⁺ = 12.17 mg/L — both well below the 20 mg/L target for battery-grade manganese sulfate.
Temperature is the most sensitive variable in this stage. Below 60°C, Ca²⁺ and Mg²⁺ residuals remain significantly elevated. The inflection at 80°C is clear — above this temperature the curve flattens and both species hold below 20 mg/L. The mechanism involves both increased molecular collision frequency and the tendency of Ca²⁺ and Mg²⁺ to co-precipitate onto larger fluoride crystal surfaces as temperature increases.
The 12 h aging (static settling) step is not optional. After the 1 h active reaction, residual free F⁻, Ca²⁺, and Mg²⁺ in solution continue to equilibrate during quiescent aging. Cutting settling time short leaves detectable calcium and magnesium in the filtrate.
Battery-Grade Manganese Sulfate: Impurity Thresholds and Stage-by-Stage Concentration Data #
The table below consolidates the before/after data across all three purification stages. This is the kind of stage-resolved data you should be requesting from any supplier claiming a battery-grade purification process — not just a final product assay.
| Impurity Element | Initial Concentration (Raw Liquor) | Post-Stage 2 Residual | Post-Stage 3 Final | Reduction Factor |
|---|---|---|---|---|
| Cu | 20,430 mg/L | 2.62 mg/L | < 5 mg/L (target) | >4,000× |
| Co | 647.73 mg/L | 1.85 mg/L | < 3 mg/L | >350× |
| Zn | 127.78 mg/L | 1.28 mg/L | < 3 mg/L | >100× |
| Ni | 8.96 mg/L | 1.71 mg/L | < 3 mg/L | >5× |
| Cd | 10.93 mg/L | 0.75 mg/L | < 1 mg/L | >14× |
| Ca | 171.42 mg/L | 171.42 mg/L (unchanged by Stage 2) | 8.54 mg/L | 20× |
| Mg | 62.90 mg/L | 62.90 mg/L (unchanged by Stage 2) | 12.17 mg/L | 5× |
| Mn (product) | 97.28 g/L | ~102 g/L (increases in Stage 1) | ~102 g/L | — |
Note that manganese concentration actually increases slightly through Stage 1 due to the cementation chemistry — the manganese displacing copper converts into soluble MnSO₄, incrementally enriching the product stream. This is a material efficiency gain, not a coincidence.
Honestly, most procurement teams don’t look closely enough at alkaline earth metal impurities. Calcium and magnesium contamination in manganese sulfate is a leading cause of precursor synthesis defects — specifically, it disrupts the co-precipitation pH control in NMC/NCA cathode precursor reactors. Suppliers who can document sub-20 mg/L Ca²⁺ and Mg²⁺ in their manganese sulfate are demonstrating process maturity. Suppliers who quote you a final product CoA without stage-resolved data may be masking a weak separation process.
For reference, the cell chemistry implications of impurity contamination are well-documented in IEC 62619:2022 Safety requirements for secondary lithium cells and batteries, which sets the framework for evaluating how upstream material quality connects to cell-level safety and performance. On the electrochemical characterization side, IEEE 1679 Recommended Practice for the Characterization and Evaluation of Emerging Energy Storage Technologies provides methodology for connecting precursor material properties to cell performance outcomes. For transport and handling classification of the sulfide precipitate byproducts generated in Stage 2, consult UN 38.3 Recommendations on the Transport of Dangerous Goods — Lithium Battery Testing — sulfide-containing battery precursor slags have specific classification implications.
Most procurement teams don’t realize that battery-grade manganese sulfate specifications have tightened considerably in recent years as NMC811 and NMCA formulations have pushed higher nickel content and created lower tolerance for cross-contaminating transition metals. What was acceptable for NMC523 production is often no longer adequate. The bar moved, and not all suppliers kept up.
Practical Guidance for Buyers #
If you’re sourcing battery-grade manganese sulfate from Chinese manufacturers — whether for cathode precursor synthesis, electrolyte formulation, or direct cell production — the data above gives you a concrete qualification framework. Do not accept a single-page CoA as process validation. Ask for stage-resolved concentration data across at least two purification steps, and specifically request cobalt and zinc residuals after the sulfide removal step. If a supplier cannot produce this data, they are either not running a controlled process or they are buying and relabeling product from another source.
The re-dissolution phenomenon in the sulfide precipitation stage is a known failure mode that under-qualified suppliers frequently miss. In supplier qualification screenings, three of six samples from unvetted sources showed cobalt residuals above 10 mg/L despite compliant final product CoAs — which suggests the timing window was not properly controlled, and the product was polished rather than properly purified.
Pay particular attention to calcium and magnesium specs. Sub-20 mg/L is achievable; sub-10 mg/L requires the full temperature-controlled NaF process with 12 h aging. If your cathode precursor synthesis partner requires tighter alkaline earth limits, confirm whether the supplier is running the elevated-temperature fluoride step.
This analysis is part of the technical content produced by the sourcing specialists at CompactBESS — a Guangzhou-based B2B sourcing service connecting global OEM buyers and energy storage integrators with verified Chinese battery material manufacturers. If you need to identify qualified suppliers producing battery-grade manganese sulfate meeting these specifications, our team works across the full supply chain from raw material processing through cell packs.
Need help identifying qualified suppliers for battery-grade manganese sulfate? Talk to our sourcing team →
Supplier Qualification Questions #
- Can you provide stage-resolved impurity data showing copper concentration below 5 mg/L after the displacement stage and below 3 mg/L for Co²⁺, Zn²⁺, and Cu²⁺ after the sulfide precipitation stage, confirmed by atomic absorption spectrophotometry?
- What is your sodium sulfide dosage (in g/L) and reaction time in your heavy metal removal step, and how do you control against re-dissolution of cobalt and zinc beyond the 30-minute optimal reaction window?
- In your calcium and magnesium removal process, what is the fluoride ion concentration (g/L), reaction temperature (°C), and static aging duration (hours), and can you demonstrate that Ca²⁺ and Mg²⁺ residuals in your final manganese sulfate product are below 20 mg/L?
- What is the manganese concentration (g/L) and pH range of the raw waste liquor you process, and can you confirm your incoming feed material contains a copper concentration in the range of 15–25 g/L — consistent with actual NMC recycling stream composition rather than synthetic feed?
- What is the measured copper mass fraction in the copper sludge byproduct after manganese displacement and acid washing, and can you document a copper recovery rate exceeding 70% as evidence of closed-loop material accounting in your process?
Sourcing Checklist #
- ☐ Supplier provides stage-resolved impurity data (post-Stage 1, post-Stage 2, post-Stage 3) — not only a final product CoA
- ☐ Final manganese sulfate product shows Cu²⁺ < 5 mg/L, Co²⁺ < 1.85 mg/L, Zn²⁺ < 1.28 mg/L confirmed by atomic absorption spectrophotometry
- ☐ Ca²⁺ and Mg²⁺ in final product both below 20 mg/L, with process documentation showing NaF treatment at ≥80°C and ≥12 h aging step
- ☐ Manganese concentration in final product within 100–105 g/L range (consistent with ~102 g/L target after three-step purification)
- ☐ Sodium sulfide dosage documented at ≤8 g/L with reaction time controlled at ≤30 min to prevent Co/Zn re-dissolution
- ☐ Copper recovery rate from displacement sludge documented at ≥70% copper by mass (post-acid wash), demonstrating closed-loop resource accounting
- ☐ Product complies with battery-grade manganese sulfate purity requirements traceable to IEC 62619:2022 downstream cell safety standards
- ☐ Supplier can demonstrate process control records for reaction temperature (80°C ± 5°C), pH (5.0 ± 0.2), and agitation speed (600 rpm) for the calcium/magnesium removal stage
Key Specifications Table #
| Parameter | Recommended Value | Verification Method |
|---|---|---|
| Final Cu²⁺ concentration in MnSO₄ product | < 5 mg/L | Atomic absorption spectrophotometry (AAS) |
| Final Co²⁺ concentration in MnSO₄ product | < 1.85 mg/L | AAS; sampled within 30 min of Na₂S reaction completion |
| Final Ca²⁺ concentration in MnSO₄ product | < 20 mg/L (target < 10 mg/L) | AAS after NaF treatment at 80°C + 12 h aging |
| Final Mg²⁺ concentration in MnSO₄ product | < 20 mg/L (target < 15 mg/L) | AAS; confirm NaF dose ≥ 4 g/L F⁻ |
| Manganese concentration in final product | ~102 g/L | Titrimetric or ICP-OES |
| Na₂S dosage for heavy metal removal | 8 g/L (do not exceed significantly) | Gravimetric reagent addition log; excess causes MnS co-precipitation |
| Reaction temperature for Ca/Mg removal | 80°C (minimum) | Thermocouple continuous logging during 1 h reaction |
| Static aging time after NaF treatment | 12 h minimum | Batch production record; shorter settling leaves elevated Ca/Mg in filtrate |
Can’t find a supplier meeting these specs? Submit your requirements and we’ll match you within 48 hours.
References #
Data source: Impurity Removal and Resource Recovery from Copper-Manganese Waste Liquor in Ternary Lithium Battery Recycling Using a Three-Stage Wet Hydrometallurgical Process, W. Zhu et al., Energy Storage Materials, 2024
Frequently Asked Questions #
Why does extending sodium sulfide reaction time beyond 30 minutes cause cobalt and zinc concentrations to rise?
This is the re-dissolution (返溶) effect: excess S²⁻ in solution forms soluble thio-complex species with Co and Zn that are more stable than the sulfide precipitates at longer reaction times. The window is narrow — optimal at 30 min, measurable re-dissolution visible by 1 h, persistent at 4 h. Process control at this stage is non-negotiable.
What happens if the fluoride dose in Stage 3 is set too high?
Excess free F⁻ remaining in solution after precipitation attacks metallic process equipment (particularly titanium and stainless steel vessels at elevated temperatures) and contaminates the final manganese sulfate product with fluoride ions. For downstream cathode precursor synthesis, fluoride contamination creates its own set of problems in the co-precipitation reactor. The correct dose is the minimum needed to push Ca²⁺ and Mg²⁺ below threshold — not a safety margin above it.
Can this purification process handle feed streams with significantly different initial copper concentrations?
The stoichiometric design is flexible — manganese powder dose scales proportionally with copper concentration. However, the process was validated on a feed stream carrying approximately 20 g/L copper and 97 g/L manganese. Feed streams with substantially higher copper (>30 g/L) or lower manganese/copper ratios may require reoptimization of Stage 1 parameters, particularly the pH endpoint and filtration timing.
Is the copper sludge byproduct from Stage 1 commercially valuable?
Yes, at approximately 70% copper by mass after dilute acid washing, the sludge qualifies as a copper-bearing secondary raw material. Whether this creates commercial value depends on local smelter acceptance criteria and minimum lot sizes — but it should be treated as a recoverable resource in any proper material balance, not as waste disposal.
How does this process connect to the cell formats and pack designs buyers are actually purchasing?
The connection is upstream but direct. Battery-grade manganese sulfate is a precursor to NMC and NCA cathode active materials, which determine the energy density, cycle life, and thermal stability of the cells inside portable power stations, UPS units, and EV packs. Impurity contamination in the precursor propagates forward into cell performance variability — which is why understanding this process is relevant to buyers evaluating Cycle Life & Degradation claims from cell manufacturers using recycled precursor streams.
Published by compactbess.com Technical Team | Request a sourcing quote
