Steel’s hottest water fight: the chemical program keeping cooling towers clean

Steel plants run some of the most punishing water circuits in industry, and open (contact) cooling towers are the workhorses. In these systems, evaporation and blowdown can consume 80–95% of makeup water unless the chemistry is controlled (prochemtech.com). The program below coordinates scale inhibitors, corrosion inhibitors, and an aggressive biocide regime to hold the Langelier Saturation Index (LSI, a scaling tendency indicator) near zero (LSI≈0.0–+0.5) and away from the “severe scaling” range (watertechonline.com). The aim is to run at the highest feasible cycles‑of‑concentration (COC, the ratio of dissolved solids in recirculating water to makeup), because higher COC directly cuts blowdown—raising COC from ~2.2 to 10 cut blowdown in one case from ~8.08 million to ~1.08 million gal/year (prochemtech.com).

This is packaged as a comprehensive cooling-tower chemical program anchored by continuous dosing and verified by on‑line monitoring.

Scale control at high cycles

The chief deposits are calcium carbonate, with CaSO₄ or silica occasionally present. The program blends threshold/polymeric inhibitors (e.g., polyacrylates, polyphosphonates, phosphonocarboxylates) plus dispersants dosed at 1–10 mg/L to sequester Ca²⁺/Mg²⁺ and keep crystals non‑adherent. Practically, that means a coordinated feed of scale inhibitors and dispersant chemistry that sustains high saturation without hard scale. A phosphonate+polymer program on hard makeup (e.g., 150 mg/L as CaCO₃ hardness) typically achieves ~2–4× COC (prochemtech.com); with softened makeup or stronger inhibitors, one case reached COC≈10 with no scale (prochemtech.com).

Dosing is tuned to conductivity and saturation indices: for makeup at 100 mg/L hardness and 150 mg/L alkalinity, COC=5 drives hardness to 500 mg/L and alkalinity to 750 mg/L. Without inhibitors, this scenario runs near LSI≈+2 (very scale‑prone), and industry guidance flags LSI>1 as “extreme scaling” (watertechonline.com); with polymeric inhibitors, adherent deposits are averted. Benchmarks call for orthophosphate around 5–15 ppm plus a phosphonate/polymer blend and 0.5–2.5 ppm zinc for steel protection (chemtreat.com), delivered by an accurate chemical dosing pump. Because phosphates can promote algae in towers (chemtreat.com), the program minimizes phosphate and leans on high‑performance polymers where feasible (chemtreat.com).

Makeup conditioning upstream—clarification and softening—expands the operating window. Where hardness must be reduced, plants deploy a softener on the makeup line; where turbidity is the driver, a clarifier can trim suspended solids before they concentrate in the tower.

Corrosion control and metallurgy

Targets for open cooling loops are ≤1 mil per year (mpy; 1 mil = 0.001 inch) steel loss, with many programs driving below 0.5 mpy. Inhibitors include orthophosphate, zinc (sacrificial passivation), molybdate, nitrate or silicate, plus trace azoles for copper alloys. A representative feed is ~1–3 ppm zinc and/or nitrite with 5–10 ppm orthophosphate‑equivalent, upgraded by film‑forming polyacrylate (chemtreat.com). One documented case—an advanced silicate/phosphonate chemistry with softened makeup—achieved 0.25–0.5 mpy on steel at 10× COC (prochemtech.com); by contrast, systems lacking inhibitors can see 5–10 mpy or worse. Real‑time coupons or probes verify rates in‑situ, with dose adjustments to keep steel loss well below 1 mpy and copper/brass near 0.01–0.2 mpy—a brief achievable via modern corrosion inhibitors.

Because circulating water is aerated and near‑neutral (pH ~7.5–8.5), cathodic corrosion is readily sustained. The program maintains slightly alkaline pH (~8.0–8.5) to support phosphate passivation while avoiding “zinc white rust” (galvanic coating attack above pH ~8.2; prochemtech.com). If molybdate or similar is used, feeds are phased to minimize any adverse effect on biocide kill (bohrium.com). Tower basin concrete, if present, is protected by lining or by maintaining slightly higher pH with silicate so any leached Ca becomes harmless carbonate.

For non‑contact (closed‑loop) systems, the practice shifts to pH 8.5–9.0 with amine or phosphate dosing and nitrite/zinc for carbon steel; these closed loops run isolated corrosion programs and do not share biocide feed. Where closed loops exchange heat with the tower, procedures prevent cross‑contamination—supported by dedicated closed-loop chemicals.

Biological fouling and biocide regimen

Warm recirculating towers are ideal habitats for bacteria, fungi, algae and protozoa. Biofilms are especially damaging: slime layers insulate and trap debris, cutting heat transfer and choking flow areas—often more effectively than other deposits (chemtreat.com; wateronline.com). The program runs a multi‑tiered biocide plan: a continuous oxidizing residual (1–3 ppm free chlorine or stabilized bromine) to police planktonic bacteria, periodic oxidizing shocks (10–20+ ppm chlorine or chlorine dioxide) to penetrate films, and regular pulses of non‑oxidizing agents (glutaraldehyde, DBNPA, isothiazolinones at ~5–20 ppm) that inactivate sessile cells and fungi. Lab tests show major classes (chlorine, bromine, isothiazolinone) can eradicate amoebic cysts/bacteria in ∼4 hours (bohrium.com), and ozone has reported ~99% control versus ~89% for chlorine alone in a seawater test (eeer.org).

The biocide feed is coordinated with inhibitors because molybdate/phosphates can lengthen kill times (bohrium.com), so pulses are alternated and de‑synchronized. Legionella prevention is part of the design: research notes the organism cannot survive above pH≈9.2 (prochemtech.com). The program monitors heterotrophic plate count (HPC) and ATP, holds HPC <10^3 CFU/mL (many programs aim <500–1000 CFU/mL), achieves >99.9% kill during shock, and maintains >0.5 ppm sanitizer between shocks—powered by modern biocide formulations. If ozone or chlorine dioxide are used, off‑gassing and safety are managed.

Solid and oil control in side‑streams

Steel operations introduce suspended solids (mill scale, grit) and hydrocarbons to towers, which seed scale and biofilm. Mechanical polishing runs in parallel: side‑stream filtration continuously treats 5–10% of recirculating flow, with automatic screen filters intercepting ≥10 µm particles and multi‑media sand capturing finer silt (chemtreat.com). DOE guidance says side‑stream filtration boosts heat transfer and can increase allowable COC (energy.gov). Typical design handles ~10–20% of recirculation with self‑cleaning backwash, pairing an automatic screen with a sand media filter or a downstream cartridge filter where finer polishing is required.

Oil control runs alongside solids removal: a coalescing plate pack or grease trap removes free oil, with basin oil content kept <2–5 mg/L to avoid spreading sheens; defoamers and possibly a demulsifier help knock out emulsions. Advanced steel‑mill trains demonstrate effluent ≤1 mg/L oil after clarifiers (patents.google.com), a level compatible with a compact oil removal unit. Side‑stream filtration also reduces basin sludge; in DOE testing, it doubled the interval between required tower cleanings (energy.gov). The operating plan still schedules quarterly drain‑and‑clean cycles.

Makeup conditioning is part of the regime: if raw iron exceeds 0.1 ppm, pre‑oxidation/filtration is triggered because >0.1 ppm Fe can precipitate black oxide and seed copper corrosion (watertechonline.com). High silica (>100 ppm) limits cycles and may require softened makeup or acid addition (silica is more soluble at low pH).

Monitoring, blowdown, and compliance

Continuous monitoring tracks conductivity (for cycles), pH, oxidant residual/ORP, and on‑line corrosion probes. Targets include COC ≈4–6×, near‑neutral LSI (Ryznar ~6; the Ryznar Stability Index is a related indicator), steel corrosion ≪1 mpy, and basin TSS <10 mg/L. Quarterly shutdown inspections—via infrared thermography or ΔT checks—verify minimal fouling factors. The LSI target band of 0.0–+0.5 stays below the “severe scaling” range, with LSI>1 flagged as “extreme scaling” (watertechonline.com).

Water savings are material: pushing COC from ~2.2 to 10 reduced blowdown 8.08→1.08 million gal/year in one 1000‑ton case (prochemtech.com) and saved ~$70.6K annually (prochemtech.com). Another analysis annualized ~5 million gallons/year makeup reduction (from ~18 to 13 gpm makeup) (prochemtech.com). Our cooling tower (modeled at X gpm flow) stands to save millions of gallons and >$50,000/year in utility cost.

All depressurized blowdown is monitored to meet discharge limits (e.g., TSS, oil, BOD restrictions), with Indonesian regulations (e.g., Government Reg. No. 82/2001) limiting TSS and oil to low mg/L levels. Where needed, post‑treatment is applied—settling or an ultrafiltration step on blowdown—to align with those limits. Program filtration goals (≤20 ppm solids, ≤1 ppm oil; patents.google.com) comfortably meet the thresholds.

Integrated program outcomes

The combined chemistry‑plus‑mechanical approach—(1) optimized cycles via high‑performance scale control, (2) targeted passivation to achieve <0.5 mpy on steel, (3) coordinated oxidizing and non‑oxidizing biocides, and (4) side‑stream filtration with oil separation—keeps the tower and exchangers clean. In pilot tests with heavy algal loading, oxidizing choices were decisive: ozone ~99% kill, chlorine dioxide ~97%, chlorine ~89% (eeer.org); lab work also shows major biocide classes eradicate amoebic cysts/bacteria in ∼4 hours (bohrium.com). With proper control, biofilm fouling of exchangers is negligible, whereas unprotected systems can lose 10–20% thermal efficiency to biofilms (wateronline.com). A conservative rule of thumb—1 mm of scale can cut transfer by ~5–10%—drives the program’s aim to hold fouling below 0.1 mm equivalent permanently.

Case histories back the economics: advanced treatment in a similar plant saved ~5 million gallons/year and ~$70.6K by achieving ~10× COC (prochemtech.com; prochemtech.com). Filtration does not replace chemicals; it enhances them by removing particles that otherwise blunt inhibitors and biocides (energy.gov; chemtreat.com). The result is a tower that runs cleaner, longer, and with fewer shutdowns.