Turning scrubber slurry into drywall: the hydrocyclones and vacuum belts behind FGD gypsum’s big bet

Flue gas desulfurization (FGD — the process coal plants use to scrub SO₂) now yields one of the construction industry’s most reliable gypsum sources. By 2020, synthetic gypsum output clocked in at roughly 250–260 million tonnes per year, with Asia producing ~55%, Europe 22%, and North America 18% (ScienceDirect). For context, a single 450 MW unit burning 0.8%-S coal about 6,000 hours a year can generate ≈50,000 tonnes of FGD gypsum, the second‑largest coal‑combustion byproduct after fly ash (ZKG) (ZKG).

That supply feeds a drywall market that already leans synthetic. In the U.S., nearly half of the gypsum used in drywall is FGD‑derived (GlobeNewswire). Back in 2006, the U.S. consumed ~41.6 Mt of gypsum — 91% of it in wallboard (Power Engineering).

Today the synthetic gypsum market is forecast to grow from an estimated US$2.3 billion in 2022 to ~$3.2 billion by 2030 (≈4.3% CAGR), driven by construction demand and scrubber incentives in coal‑heavy regions such as China and India (GlobeNewswire) (GlobeNewswire) (GlobeNewswire). The wallboard/drywall segment dominates (≈50–65% of FGD gypsum use), prized for consistent quality and purity (GlobeNewswire) (Power Engineering) (GlobeNewswire).

FGD gypsum volumes and regional shifts

Western supply is expected to peak by ~2030 and then decline as coal plants retire across Europe, Japan, and the U.S. (Rhewum). Europe already uses ~4 Mt/y of FGD gypsum, covering ≈50% of German gypsum demand (Rhewum). In Indonesia, future supply could grow with planned coal capacity additions (RUPTL 2019–2028 envisages +27 GW), though historically lax SO₂ limits (e.g., 750 mg/Nm³ in older regulations) curbed scrubber adoption (GlobeNewswire) (Mongabay).

Even so, new Indonesian units are installing wet‑limestone FGD. The Sumsel 8 project (2×660 MW) is a recent example, and plants of that size produce tens of thousands of tonnes of gypsum per year (ESDM).

Two‑stage dewatering train (hydrocyclone + vacuum belt)

Wet‑limestone scrubbers discharge a dilute gypsum slurry at about 10–15 wt% solids (wt% = weight percent), which must be concentrated to a dry, handleable filter cake. Industry practice uses a two‑step solid–liquid separation: hydrocyclone pre‑thickening followed by vacuum filtration (typically horizontal vacuum belt filters or centrifuges) (bulk‑online) (Filtration + Separation).

Hydrocyclones, which use centrifugal force to separate solids from liquid, typically lift solids from ~15% to roughly 35–60% before filtration; some configurations report ~50–60% and even ~35–70% solids, and the stage also classifies out very fine “dead” particles (bulk‑online) (Filtration + Separation).

Filter performance and equipment choices

Horizontal vacuum belt filters then take the thickened slurry to final dryness. Modern belts (including monofilament mesh that improves airflow and cake release) offer filter areas up to 100–200 m² and can push very high throughputs; one belt‑filter train can handle on the order of 80–90 tph (tonnes per hour) of wet gypsum slurry (bulk‑online). At Germany’s Neurath plant, two 35×3.5 m belts each process 81 tph; such belts remove >90% of the water and deliver filtrate with ≪0.5% solids (bulk‑online).

The result is a filter cake typically under 10% moisture; tests at the Belchatov lignite plant consistently showed <10% residual moisture, with state‑of‑the‑art belts achieving ~8–10% (bulk‑online) (bulk‑online).

Equipment selection hinges on dryness targets and feed variability. For very high cake‑dryness and fluctuating feeds, vertical basket (peeler) centrifuges are often chosen; they can exceed 90% solids, are easier to clean, and operate continuously. Belt filters win when throughput is high and space permits, trading a larger footprint for lower capex per ton; investment sizes reach ~100 m² belts and >50 m wide rolls (Filtration + Separation) (bulk‑online) (Filtration + Separation). By comparison, simple drum filters/drum centrifuges generally leave higher moisture and see less use (bulk‑online).

Wallboard‑grade purity requirements

Drywall plants demand a clean, consistent feed. Gypsum purity must be >92–96% CaSO₄·2H₂O (calcium sulfate dihydrate); that means full oxidation of sulfite to sulfate (sufficient air) and high‑purity limestone reagent (≥93.5% CaCO₃). U.S. wallboard tests show ~96% gypsum content by TGA (thermogravimetric analysis, a heating‑based method) on FGD feed, and FGD gypsum can exceed the purity of natural gypsum when well controlled (Power Engineering) (Power Engineering).

Moisture in the raw cake should be ≤10% for efficient calcining; most calcining plants expect 7–10%. Too wet a cake causes handling problems such as bridge arching in silos (Power Engineering) (Filtration + Separation).

Chloride and other soluble salts are the most critical impurities. Wallboard guidelines typically limit Cl⁻ to ≈0.01 wt% (100 ppm). Untreated scrubber slurries can contain chloride in the tens of thousands of ppm, and even at 10% moisture the solids can carry ~0.5–1% chloride if not washed (Power Engineering) (Filtration + Separation).

Other impurities to control include heavy metals (Pb, Hg, As, etc.) and unreacted limestone (CaCO₃). Calcium sulfite (incomplete oxidation) must be kept under 5 mol% of total CaSO₄/CaSO₃. Crystal morphology matters too: well‑formed dihydrate crystals (often prismatic, mean ~40–90 μm) dewater and calcine better than fines or needle‑like sulfite phases; age‑soak or seed recycle can help (Filtration + Separation) (Filtration + Separation).

Example: U.S. drywall standards specify calcined gypsum with ≥92% CaSO₄·2H₂O and <100 ppm chloride; California kiln specs are similar. Utilities targeting the wallboard market commonly aim for ~95% gypsum purity and cake moisture <8–10% (Power Engineering) (Power Engineering).

Washing and purification to meet specs

Getting to wallboard‑grade usually means an aggressive wash. A freshwater rinse on the vacuum belt dissolves Ca/Mg/Na chlorides and can drop chloride from about 1% in slurry to under 100 ppm in cake. Operators verify with simple bench tests (rinse an aliquot, titrate) and adjust wash water as needed (Power Engineering) (Power Engineering) (Filtration + Separation).

Some plants go further with an acid wash at pH ~2–5 to dissolve residual CaCO₃ and purge certain Fe/Al impurities. One patented method routes the slurry (after optional magnetic separation) through an acid bath at pH<5, then neutralizes and refilters; the approach can lighten color and remove up to >99% of trace metals depending on chemistry (US20110044883A1) (US20110044883A1) (US20110044883A1). Precise reagent addition for this step is typically controlled with a dosing pump to maintain the target pH.

Magnetic separation (HGMS — high‑gradient magnetic separation) can polish out ferrous particles ahead of washing, particularly if limestone contains ferric minerals that darken product; the cited patent integrates a magnetic trap before the acid bath (US20110044883A1).

For extreme purity or dryness targets, operators may re‑slurry and run a second filter or a centrifuge stage; in some cases, a final thermal dryer or calciner follows. Vendors offer continuous filters with internal spray zones to optimize wash efficiency (Filtration + Separation) (bulk‑online).

Filtrate handling and quality control

The rinse filtrate is settled/neutralized and recycled to the process; a dedicated clarifier is commonly applied for this duty. Quality checks — chloride, sulfate, carbonate, heavy metals — are run by lab assay or in‑line measurement to confirm the clean, dry cake (6–10% moisture) meets spec (Power Engineering).

Indonesia’s emerging FGD supply

Indonesia’s construction market is growing, and local wallboard makers currently lean on natural gypsum and imports. As more units add FGD — like Sumsel 8’s wet‑limestone system — synthetic gypsum volumes could rise, provided product meets strict purity specs (ESDM). Market potential is clear, but without stricter SO₂ standards or incentives, byproducts risk being handled as waste, not feedstock (Mongabay) (GlobeNewswire).

Outlook and market trajectory

With hydrocyclone thickening, vacuum filtration plus effective wash (and, where needed, acid polishing or magnetic separation), modern FGD systems reliably hit wallboard specs — >92% CaSO₄·2H₂O, <10% moisture, <0.01% Cl⁻, and minimal heavy metals (Power Engineering) (Filtration + Separation) (US20110044883A1). That aligns with a market set to grow to ~$3.2 billion by 2030 (≈4.3% CAGR) even as Western supply tightens and Europe/Japan explore recycling to offset declining FGD output (GlobeNewswire) (Rhewum).

For Indonesian utilities and regulators, the technical path is proven: install effective scrubbers and dewatering trains so gypsum becomes a saleable resource rather than a disposal problem — a shift already underway in new builds and supported by the established hydrocyclone‑plus‑vacuum‑belt playbook (bulk‑online) (Power Engineering).