UV/chlorine vs. UV/H2O2 for water reuse at Orange County Water District, CA: a pilot study

Abstract

On-site demonstration tests evaluated 1,4-dioxane removal in reverse osmosis permeate (RO permeate) at the Orange County Water District (OCWD) Advanced Water Purification Facility (AWPF). The study used a pilot-scale ultraviolet (UV) reactor to compare efficacy of UV/chloramine, UV/chloramine/H₂O₂ and UV/chloramine/HOCl advanced oxidation processes (AOPs). The UV/chloramine process did not achieve the targeted 0.5 log removal of 1,4-dioxane even at a high UV fluence of 4400 mJ cm⁻² and the 1,4-dioxane decay did not follow pseudo-first order kinetics. Oxidant concentrations as low as 2.6 mg L⁻¹ H₂O₂ or ∼1.0 mg L⁻¹ free Cl₂ (HOCl) in the RO permeate at ambient quality parameters, e.g. 2.2–2.4 mg L⁻¹ monochloramine (as Cl₂), 1.2–1.4 mg L⁻¹ dichloramine (as Cl₂), ≤0.08 mg L⁻¹ NO₂⁻ (UV/H₂O₂ AOP), ∼97.0% T, resulted in 0.56 and ≥0.51 log reduction of 1,4-dioxane with the UV/chloramine/H₂O₂ and UV/chloramine/HOCl process, respectively, while simultaneously achieving ∼2.0 log reduction of NDMA. Overall, the UV/chloramine/HOCl process showed higher removal efficiency of 1,4-dioxane than UV/chloramine/H₂O₂ for the same mass-based oxidant concentrations and process conditions. Nitrite ions present in the RO permeate and/or generated from monochloramine photolysis during the UV/AOP impacted significantly the process performance. It was confirmed through these pilot-scale tests that the removal of 1,4-dioxane in the UV/chloramine/H₂O₂ and UV/chloramine/HOCl processes can be predicted accurately with a comprehensive mechanistic kinetic model that accounts for the molecular and radical reactions that occur in the RO permeate in the UV reactor.