Hospitals Are Winning on Legionella — and Losing Copper and Zinc

Hospitals have leaned on secondary disinfection to keep Legionella in check, but the same chemicals that clean the water can quietly eat through the plumbing. In two Italian hospitals, corrosion coupons clocked carbon‑steel losses of ~0.11 mm/yr under monochloramine, ~0.14 mm/yr under chlorine dioxide, ~0.17 mm/yr under hydrogen peroxide, and ~0.14 mm/yr even when untreated (pmc.ncbi.nlm.nih.gov).

The morphologies differ — chlorine dioxide produced uniform plus ulcerating corrosion, hydrogen peroxide drove deep pitting, and monochloramine showed mostly uniform mild attack (pmc.ncbi.nlm.nih.gov; pmc.ncbi.nlm.nih.gov). Even where rates remain “satisfactory” (<0.25 mm/yr is often treated as a maintenance threshold), the choice of disinfectant changes how and where metals fail — making monitoring essential.

Chlorine residuals and copper loss

Chlorine (hypochlorite) is a strong oxidizer that accelerates copper corrosion; empirical data show copper corrosion rates rise roughly in proportion to the residual dose (pmc.ncbi.nlm.nih.gov). In one study, copper coupons in tap water lost progressively more mass as free Cl₂ increased from ~0.25 mg/L to 1 mg/L — with both corrosion rate and copper leaching climbing in step (pmc.ncbi.nlm.nih.gov).

Even low‑level chlorination (~0.25–0.5 mg/L) produced a measurable copper loss on the order of 0.01 mg/d·cm² in lab tests (pmc.ncbi.nlm.nih.gov), while shock‑level treatments (5 mg/L) drove multiple‑fold higher rates — in one immersion test, reaching hundreds of mils per year (mpy, a corrosion unit) (pmc.ncbi.nlm.nih.gov). Galvanized steel (zinc‑coated) also suffers: microscopy shows chlorine, and especially chlorine dioxide, progressively dissolving zinc, with copper surfaces turning porous as oxides form — classic dezincification on galvanized steel (pmc.ncbi.nlm.nih.gov).

Chlorine dioxide attack on copper and zinc

Chlorine dioxide (ClO₂) is strongly oxidizing and particularly aggressive to metals. At ~1 mg/L ClO₂ in lab simulations, copper converted to very porous copper oxide and galvanized steel showed intense dezincification (pmc.ncbi.nlm.nih.gov). Vertova et al. reported ClO₂ makes copper surfaces “very porous” and strips zinc from galvanized steel (pmc.ncbi.nlm.nih.gov).

One accelerated study found “severe degradation” across copper and zinc materials under 0.8 mg/L ClO₂ at elevated temperature (pmc.ncbi.nlm.nih.gov). On carbon steel, hospital coupons under chlorite treatment measured ~0.14 mm/yr — approaching commonly cited acceptable limits when expressed in hundreds of μm/yr (pmc.ncbi.nlm.nih.gov). The practical takeaway: ClO₂ control requires rigorous coupon surveillance.

Monochloramine and the ammonia factor

Monochloramine (NH₂Cl), a more stable, lower‑oxidation biocide, is generally less aggressive toward metals than chlorine or chlorine dioxide (www.sanipur.com). In Marchesi’s field work, carbon‑steel coupons under monochloramine averaged only 0.11 mm/yr — the smallest among treatments (pmc.ncbi.nlm.nih.gov).

Monochloramine does not oxidize copper as readily as chlorine, but its ammonia byproduct can foster localized pitting if nitrification takes hold. Plumbing programs therefore maintain fresh hypochlorite and proper feed ratios to minimize ammonia spikes (pmc.ncbi.nlm.nih.gov). Still, chloramines slowly corrode copper and iron and can induce biofilms if nitrification is uncontrolled (www.sanipur.com), so even these systems benefit from corrosion monitoring.

Hydrogen peroxide in complex formulations

Hydrogen peroxide (H₂O₂), often used in stabilized or “accelerated” blends, is a strong oxidant that can significantly corrode copper. Nakhaie et al. reported hospital‑grade accelerated H₂O₂ cleaners caused “considerably higher corrosion rates” on pure copper than H₂O₂ alone — evidence that routine peroxide disinfection readily dissolves copper (iopscience.iop.org).

In Marchesi’s comparison, H₂O₂ treatment produced the highest steel coupon loss (~0.17 mm/yr) (pmc.ncbi.nlm.nih.gov). On galvanized steel, peroxide likely oxidizes zinc rapidly (specific data are fewer), and residuals above 10 mg/L warrant close watch because months of exposure can produce severe pitting and oxide formation.

Other disinfection modalities

Ozone and UV have little pipeline effect; ozone lacks a lasting residual deep in the system, and UV systems add no chemicals. [Copper–silver ionization (Cu–Ag)](https://beta.co.id/en/blog/hospitals-legionella-playbook-chlorine-dioxide-vs-coppersilver-vs-monochloramine) is not a “chemical” but a metal dose; it elevates copper and silver levels and can cause galvanic corrosion of ferrous components if dissimilar metals touch (pmc.ncbi.nlm.nih.gov). Since hospital cold/hot lines rarely use bare galvanized iron, Cu–Ag is less common; if present, it accelerates rusting of steel parts (pmc.ncbi.nlm.nih.gov).

Corrosion monitoring: coupons and probes

When a chemical residual is present, routine corrosion monitoring is critical. Placing metallic coupons (strips of copper, galvanized steel, or stainless) in hazardous zones for 3–6 months, then measuring weight loss, yields direct corrosion rates (mm/yr or mpy) (pmc.ncbi.nlm.nih.gov; pmc.ncbi.nlm.nih.gov). Marchesi’s 6‑month program returned 0.11–0.17 mm/yr losses (pmc.ncbi.nlm.nih.gov), and Vertova et al. used scanning electron microscopy (SEM) to show oxide films and dezincified zones (pmc.ncbi.nlm.nih.gov).

Why it matters: chemical disinfection can push metal loss beyond design allowances; without monitoring, pinholes or leaks may surface months or years later. Corrosion also consumes disinfectant (free chlorine reacts with metal oxides) and releases iron and copper that can feed biofilms (pmc.ncbi.nlm.nih.gov). Beyond coupons, facilities deploy corrosion probes — including anodic/cathodic half‑cell meters and linear polarization resistance (LPR) probes — for real‑time indices; periodic sampling for dissolved Cu, Zn, and Fe and analyzing scale chemistry complements the picture. Tracking trends is the trigger: a rising copper concentration or increasing pitting coverage on removed coupons is a red alert.

Water safety plans (WSP) and engineering guidance (e.g., ASHRAE, AWWA) treat corrosion control as a parallel requirement: if a secondary disinfectant is added, its impact on materials should be tracked. Indonesian health guidance similarly emphasizes [Legionella risk management via a WSP](https://beta.co.id/en/blog/hospitals-are-turning-water-plans-into-hard-proof-heres-the-data-playbook-that-keeps) that includes water temperature and chemistry control (pmc.ncbi.nlm.nih.gov); while the 2019 “Pedoman Pencegahan dan Pengendalian Legionellosis” focuses on pathogen control, its principles imply corrosion monitoring as part of managing hospital water distribution systems.

Water chemistry management parameters

Maintaining favorable pH and alkalinity stabilizes metals. Copper is most stable near neutral to slightly alkaline pH; soft, acidic water (pH < 6.5) aggressively dissolves copper (www.ncbi.nlm.nih.gov). WHO guidance notes neutral‑to‑alkaline pH (7–9) and moderate hardness promote protective copper carbonate films (www.ncbi.nlm.nih.gov). For galvanized/steel, the optimal pH is often ~6.8–7.5, with one guideline recommending 6.8–7.3 for iron networks (www.ncbi.nlm.nih.gov).

Sufficient alkalinity (≥40 mg/L as CaCO₃) and a slightly positive Langelier Index (a measure of scaling tendency) promote protective calcium carbonate on steel; in practice, buffering with soda ash or lime and ensuring ≥40–60 mg/L alkalinity is advantageous (www.ncbi.nlm.nih.gov). Where orthophosphate or silicate films are appropriate, targeted dosing of a corrosion inhibitor can cut iron/copper release; WHO notes orthophosphate is widely used to prevent “red water” from iron corrosion (dose must be managed to avoid phosphate‑induced biofouling) (www.ncbi.nlm.nih.gov).

Chloride control matters: chloride promotes copper pitting and galvanic action at soldered joints. Lead/copper program experience implies keeping the chloride‑to‑sulfate mass ratio (CSR) low; guidance points to CSR < 0.5 to avoid pitting (www.researchgate.net). Where source water has high chloride (e.g., sea salt intrusion), additional inhibitors or pH adjustments are needed.

Disinfectant residuals should be optimized. Higher is not always better: aim for the minimum effective residual — approximately 0.5–1.0 mg/L Cl₂ or 2–4 mg/L ClO₂ at outlets — to limit oxidant stress while maintaining efficacy. In Marchesi’s dataset, Legionella control was achieved at ~3 mg/L monochloramine, making higher doses unnecessary (pmc.ncbi.nlm.nih.gov). Frequent flushing reduces stagnation that can locally spike oxidants. Measured chemical feed using dosing pumps supports tighter control of these setpoints.

Dissolved oxygen and temperature also influence rates: aerobic conditions speed iron corrosion, so deaerated makeup and minimizing open storage can help. Keeping hot water in “Legionella‑safe” ranges (50–55 °C) reduces dissolved oxygen; large temperature swings can stress protective films, so stable recirculation is preferred. Monitoring data should drive adjustments — e.g., if coupons show rising copper loss, nudge pH upward or refine inhibitor dose; where galvanic coupling is evident, change materials or introduce dielectric unions.

Program thresholds and response

A balanced hospital strategy accepts that chemical disinfection increases corrosion stress on copper and galvanized components but uses monitoring and chemistry control to bound the risk. Quantitatively, acceptable coupon losses might be on the order of ~0.1–0.2 mm/yr. If surface loss exceeds ~0.25 mm/yr (≈250 μm/yr), immediate action is warranted — from switching disinfectant to applying inhibitors (and, where appropriate, water conditioning). With careful control, facilities can reap antimicrobial benefits without unduly shortening plumbing life. These recommendations reflect peer‑reviewed studies and WHO/AWWA treatment guidance (pmc.ncbi.nlm.nih.gov; pmc.ncbi.nlm.nih.gov; iopscience.iop.org; pmc.ncbi.nlm.nih.gov; www.ncbi.nlm.nih.gov; www.ncbi.nlm.nih.gov).