Removal of selected emerging PPCP compounds using greater duckweed (Spirodela polyrhiza) based lab-scale free water constructed wetland

Greater duckweed (Spirodela polyrhiza) based lab-scale free water constructed wetland (CW) was employed for removing four emerging pharmaceuticals and personal care products (PPCPs) (i.e. DEET, paracetamol, caffeine and triclosan). Orthogonal design was used to test the effect of light intensity, aeration, E.coli abundance and plant biomass on the target compounds. Synthetic wastewater contaminated with the target compounds at concentration of 25 μg/L was prepared, and both batch and continuous flow experiments were conducted. Up to 100% removals were achieved for paracetamol (PAR), caffeine (CAF) and tricolsan (TCS) while the highest removal for DEET was 32.2% in batch tests. Based on orthogonal Duncan analysis, high light intensity (240 μmolmm^-2s^-1), full aeration, high plant biomass (1.00 kg/m²) and high E.coli abundance (1.0 × 10⁶ CFU/100 mL) favoured elimination of the PPCPs. Batch verification test achieved removals of 17.1%, 98.8%, 96.4% and 95.4% for DEET, PAR, CAF and TCS respectively. Continuous flow tests with CW only and CW followed by stabilization tank (CW-ST) were carried out. Final removals of the PPCP contaminants were 32.6%, 97.7%, 98.0% and 100% for DEET, PAR, CAF and TCS, respectively, by CW system alone, while 43.3%, 97.5%, 98.2% and 100%, respectively, were achieved by CW-ST system. By adding the ST tank, PPCP concentrations decreased significantly faster (p < 0.05) compared with continuous flow CW alone. In addition, after removing aerators during continuous flow CW experiments, the treatment systems presented good stability for the PPCP removals. CW-ST showed better chemical oxygen demand (COD) and total organic carbon (TOC) removals (89.3%, 91.2%, respectively) than CW only (79.4%, 85.2%, respectively). However, poor DEET removal (<50%) and high E.coli abundance (up to 1.7 log increase) in the final treated water indicated further treatment processes may be required. Statistical analysis showed significant correlations (p < 0.05) between PPCPs and water quality parameters (e.g. COD, nitrate, phosphate), and between the four PPCP compounds for the continuous flow CW and CW-ST systems. Positive results encourage further test of Greater duckweed at pilot scale CW using real wastewater.