Enhancement of total nitrogen removal through effluent recirculation and fate of PPCPs in a hybrid constructed wetland system treating urban wastewater

Abstract

The effect of effluent recirculation on the removal of total nitrogen (TN) and eight pharmaceuticals and personal care products (PPCPs) was evaluated during 9months in an experimental hybrid constructed wetland (CW) system applied in the treatment of urban wastewater. An Imhoff tank was followed by three stages of CWs (two 1.5-m vertical subsurface flow (VF) beds alternating feed-rest cycles, a 2-m horizontal (HF) and a 2-m free water surface (FWS) wetland in series). A fraction of the final effluent was recycled back to the Imhoff tank with a recirculation rate of 50% (hydraulic loading rate=0.37md). The system's performance varied throughout the study. In Period I (summer) consistently high load removal efficiencies of TN (89±5%) and a removal rate of 6.6±1.4gTNmd were exhibited. In Period II (fall), the poor performance of the FWS during the senescence of macrophytes caused a large increase in organic matter, solids and nutrient concentrations, drastically deteriorating water quality. The determination of PPCPs was conducted during this period. Recalcitrant compounds, namely sulfamethoxazole, carbamazapine, TCEP and sucralose were negligibly removed in all CWs. However, noteworthy was the ≈30% removal of sucralose in the VF wetland. Caffeine (80%) and fluoxetine (27%) showed similar elimination rates in both VF and HF units, whereas trimethoprim and DEET were significantly better removed in the VF than in the HF. The concentration of the four latter compounds showed a severe increase in the FWS, indicating possible desorption from the sediment/biomass during adverse conditions. Harvesting of the aboveground biomass in this unit returned the system's performance back to normality (Period III), achieving 77±7% TN removal despite the winter season, proving effluent recirculation as an effective strategy for TN removal in hybrid CW systems when stringent restrictions are in place.

Removal of organic carbon, nitrogen, emerging contaminants and fluorescing organic matter in different constructed wetland configurations

“…While trimethoprim, which occurred at very low concentrations in the UVF effluent, was not detected after the HF bed, the concentration of caffeine was reduced by 82% in this unit. Sulfamethoxazole was negligibly removed in all CW units (≤30%) of treatment line 2, which agrees with the poor degradation of this substance reported in the literature [8,25,27]. However, the SVF reactor showed an enhanced capacity for the removal of this antibiotic, achieving moderate removal efficiencies (64 ± 32%), as opposed to Treatment Line 2 (Fig.…”

“…Neverthless, the FWS wetland showed in general a poor performance similarly to its prior operational period, presumably related to the senescence of the plant biomass in this unit, which caused the release of solids and organic compounds from the rhizosphere to the water column [25].…”

“…The details concerning these materials are reported in the Supplementary material section (Text S1). The selection of PPCPs analyzed in this study (i.e., caffeine, trimethoprim; sulfamethoxazole, N,N-diethyl-meta-toluamide -DEET-, sucralose) was based on data presented in previous literature that rely on chemical-physical properties, occurrence data, detection frequency, availability of robust analytical methods and removal during wastewater treatments [24,25]. Detailed information about all target analytes used in this study is indicated in Table S1.…”

“…It was commissioned in 2010 and since then it has been continuously operated and monitored as a hybrid system applying increasing hydraulic loads to assess its treatment capacity [25][26][27]. During the period of this study (April to July 2016) the configuration of the treatment plant was slightly modified, including two treatment lines operating in parallel (Fig.…”

“…Feeding of the VF beds was done intermittently and simultaneously by means of pumps and distribution pipes on the top that provided about 8 pulses a day. Both wetlands were operated in a continuous mode -as opposed to previous operational periods where the two units alternated feed-rest cycles of 3.5 d [25]. An electromagnetic flow meter (SITRANS F M MAGFLO®) was installed before each wetland unit so as to assist on the follow up of the flow values entering the treatment system.…”