Overview
This bench-scale study compared orthophosphate, trimetaphosphate (a polyphosphate), and an orthophosphate-silicate blend as corrosion inhibitor alternatives for lead coupons in treated surface water. The study was motivated by the common utility practice of using blended phosphates for simultaneous iron and manganese sequestration and lead corrosion control, which is problematic because polyphosphates increase lead solubility. The orthophosphate-silicate blend was tested as a potential compromise: effective iron and manganese sequestration with less adverse effect on lead. Results showed that pH elevation was the dominant control on lead solubility in this system; trimetaphosphate substantially increased dissolved lead; and the silicate blend produced a smaller, partly non-significant increase in total lead but a detectable increase in dissolved lead relative to orthophosphate alone.
Key numbers
Baseline (conditioning phase prior to treatment comparisons, lead coupon leachate, µg/L):
- Median dissolved lead: 23.5 µg/L (IQR: 10.1 to 86.5 µg/L)
- Median total lead: 98.8 µg/L (IQR: 49.1 to 205.2 µg/L)
Treatment effects (geometric mean relative to orthophosphate-only control):
- Trimetaphosphate: total lead +22 percent (95% CI: +11 to +35 percent); dissolved lead +54 percent (95% CI: +25 to +91 percent) — increase in both fractions
- Orthophosphate-silicate blend: total lead +8 percent (not statistically significant); dissolved lead +30 percent (95% CI: +5 to +61 percent)
- pH 9.5 versus 7.5: total lead -24 percent; dissolved lead -34 percent (95% CI: -15 to -48 percent) — protective
- Iron and manganese co-present (versus absent): total lead +41 percent; dissolved lead -44 percent (adsorption of dissolved lead to iron and manganese colloids)
Health Canada regulatory context: MAC for lead lowered to 5 µg/L in 2019.
Methods (brief)
Bench-scale glass jar system with 52 lead coupons. Treated water from Halifax surface water plant. Five water changes per week at 24-hour and 72-hour stagnation cycles. ICP-MS (Standard Methods 3125 and 3030). Dissolved samples filtered at 0.2 µm. Mixed effects models with repeated measures structure. Corrosion product characterization by X-ray diffraction identified cerussite, hydrocerussite, and hydroxypyromorphite.
Evidence fitness
Supports: establishing that polyphosphate sequestrants increase lead solubility and are unsuitable as sole corrosion inhibitors in lead-containing plumbing; demonstrating that the orthophosphate-silicate blend offers modest improvement over trimetaphosphate for total lead while still increasing dissolved lead; confirming pH as the dominant solubility control in Atlantic Canada source water.
Does not support: extrapolation to field-scale treatment plants (bench-scale static conditions differ from distribution system hydraulics and biofilm-pipe scale interactions); applying these results to other source water chemistries without bench-scale replication; substituting silicate for orthophosphate in regulatory compliance decisions without full-scale validation.
Limitations
Bench-scale static systems do not replicate distribution system hydraulics, biofilm, or varying flow patterns. Lead coupons have a fixed surface area that differs from actual pipe geometry in distribution networks. The increase in dissolved lead fraction under silicate treatment, even when total lead is not significantly elevated, is a concern for compliance measurements that use total lead rather than dissolved lead as the basis. The study used a single Atlantic Canada source water; utilities with different alkalinity, hardness, and NOM profiles would need to conduct site-specific bench testing before adopting the silicate blend.
Related evidence
Update history
No substantive edit history is available in this build. The full commit record is available in git.