N-nitrosodimethylamine (NDMA) formation at an indirect potable reuse facility

Full-scale experiments to evaluate N-nitrosodimethylamine (NDMA) formation and attenuation were performed within an advanced indirect potable reuse (IPR) treatment system, which includes, sequentially: chloramination for membrane fouling control, microfiltration (MF), reverse osmosis (RO), ultraviolet irradiation with hydrogen peroxide (UV/Hinf2/infOinf2/inf), final chloramination, and pH stabilization. Results of the study demonstrate that while RO does effectively remove the vast majority of NDMA precursors, RO permeate can still contain significant concentrations of NDMA precursors resulting in additional NDMA formation during chloramination. Thus, it is possible for this advanced treatment system to produce water with NDMA levels higher than regional requirements for potable applications (10ng/L). The presence of Hinf2/infOinf2/inf during UV oxidation reduced NDMA photolysis efficiency and increased NDMA formation (~22ng/L) during the secondary chloramination and lime stabilization. This is likely due to formation of UV/Hinf2/infOinf2/inf degradation by-products with higher NDMA formation rate than the parent compounds. However, this effect was diminished with higher UV doses. Bench-scale experiments confirmed an enhanced NDMA formation during chloramination after UV/Hinf2/infOinf2/inf treatment of dimethylformamide, a compound detected in RO permeate and used as model precursor in this study. The effect of pre-ozonation for membrane fouling control on NDMA formation was also evaluated at pilot- (ozone-MF-RO) and bench-scale. Relatively large NDMA formation (117-227ng/L) occurred through ozone application that was dose dependent, whereas chloramination under typical dosages and contact times of IPR systems resulted in only a relatively small increase of NDMA (~20ng/L). Thus, this research shows that NDMA formation within a potable water reuse facility can be challenging and must be carefully evaluated and controlled.