Sorption of biocides, triazine and phenylurea herbicides, and UV-filters onto secondary sludge

The sludge-water distribution of a total of 41 organic micropollutants (9 phenylurea herbicides, 11 triazines, 16 biocides and 5 UV-filters) was investigated in laboratory batch experiments with fresh secondary sludge taken from a municipal WWTP. Sorption kinetics as well as sorption isotherms were examined by analyzing the compound concentration in the aqueous and solid phase for mass balance control and quality assurance. The sorption kinetic experiments revealed a sorption equilibrium time of <2 h and adverse effects of sodium azide on the sludge-water distribution of several compounds. Sorption isotherms were constructed for 6 different spiking levels spanning 3 orders of magnitude (100 ng L−1–30,000 ng L−1) and were well described by the Freundlich model. For some compounds non-linear sorption with Freundlich exponents n < 1 revealed a decreased sorption affinity to the sludge flocs with increasing aqueous phase concentration. Therefore, sludge-water distribution coefficients (Kd,sec) were calculated from the isotherm data for a constant concentration level of 1 μg L−1. Based on the sludge dry weight (dw), the Kd,sec values of phenylurea herbicides ranged from 9 L kgdw sludge−1 (isoproturon) to 320 L kgdw sludge−1 (neburon), those of triazines from 5 L kgdw sludge−1 (atrazine) to 190 L kgdw sludge−1 (terbutryn), those of biocides from 10 L kgdw sludge−1 (N,N-dimethyl-N′-p-tolylsulfamide) to 40,000 L kgdw sludge−1 (triclocarban) and those of UV-filters from 9 L kgdw sludge−1 (phenylbenzimidazole sulfonic acid) to 720 L kgdw sludge−1 (benzophenone-3). For most compounds Kd,sec values were below 500 L kgdw sludge−1 and thus removal in WWTPs by the withdrawal of excess sludge is expected to be negligible (<10%) except for the biocides triclocarban (80–95%), triclosan (55–85%), chlorophene (30–60%), imazalil (25–55%) and fenpropimorph (15–40%) as well as the UV-filter benzophenone-3 (5–20%). A simple linear free-energy relationship (LFER) approach using the logarithmized octanol–water partition coefficient log KOW as single descriptor is discussed for a rough classification of nonionic compounds regarding their potential removal in WWTPs by sorption.

► The sludge-water distribution of 41 organic micropollutants has been examined.
► Sodium azide was inappropriate for deactivation of fresh sludge.
► Sorption equilibria were reached in less than 2 h.
► Sorption isotherms were well described by the Freundlich model.
► Sorption coefficients of nonionic compounds correlated with logKOW values.