In recent years, much interest has been given to presence of emerging organic micropollutants in municipal wastewater, some of which are endocrine disruptors, toxic, or carcinogenic. Much less attention has been paid to their presence and fate in decentralized greywater systems.

This research investigated the occurrence and dynamics of micropollutants, originating from personal care products, in greywater, and developed post-treatment advanced oxidation process (AOP) for their removal. The developed AOP uses combined vacuum-UV/UV-C irradiation. In aqueous solutions, vacuum-UV generates hydroxyl radicals in-situ via homolysis of water. Such radicals are strong non-selective oxidizing agents that can degrade micropollutants.

Diurnal and seasonal patterns of greywater discharges and micropollutants concentrations were monitored at source. The monitoring was achieved using custom-built automatic sampling system designed to overcome the erratic GW generation behavior which is a characteristic of small collection systems. As expected, of the micropollutants concentrations exhibited high variability from below detection limit (0.5µgL^-1) to over a hundred µgL^-1. The most frequently detected micropollotants were methylparaben (preservative), galaxolide (fragrance) and oxybenzone (UV-filter), which are common ingredients in many PCPs. Their daily loads were 2,840, 1,887 and 728 µg p^-1 d^-1, respectively. Considering these high loads and that both galaxolide and oxybenzone are toxic and endocrine disruptive, they may pose environmental concern. Triclosan was always detected but at lower concentrations. Due to its limited biodegradability and persistence in the environment it was selected as a model micropollutant.

Triclosan degradation kinetics and identification of intermediate products were investigated under 254 nm and under combined 254/185 nm irradiation both in dry thin films and in aqueous solutions. In the latter, degradation was faster under combined 254/185 nm radiation, although 185nm radiation accounted for only 4% of total UV light intensity. This enhancement diminished in presence of •OH scavenger (methanol) and in dry films, indicating that it was mainly a result of oxidation by generated radicals rather than direct photolysis under vacuum-UV. The same phenomena was observed when triclosan was dissolved in treated greywater, resulting from presence of •OH scavengers (particulate and dissolved organic matter and carbonate ions). The most significant improvement of the vacuum-UV removal of triclosan in greywater was achieved when particulate matter was removed (filtration).

Three main transformation products were identified during triclosan photooxidation: 2,8-dichlorodibenzo-p-dioxin, 5-chloro-2-(4-chlorophenoxy)phenol and 2-hydroxy-8-chlorodibenzodioxin. Under vacuum-UV irradiation (254/185 and 172 nm) both maximum concentrations and persistence of these intermediates was lower than under 254nm radiation alone. As a result, toxicity of triclosan solution irradiated by combined lamp was lower than that irradiated by UV-C light only.

Complementary experiments with two additional micropollutants (oxybenzone and tonalide) under 254, 254/185 and 172 nm, revealed similar results: enhanced photooxidation under vacuum-UV compared with slower (tonalide) or no (oxybenzone) removal under 254 nm.

In summary, AOP based on combined vacuum-UV and UV-C irradiation has promising potential for both micropollutants removal and pathogens inactivation in decentralized water treatment systems. The main merit of this process is that it does not require addition of chemicals/photocatalyst, simplifying its operation and maintenance.