Search

Rinsing tanks of crop sprayers produce significant volumes of Cu-rich Bordeaux mixture effluents (BME) that can be treated by rhizofiltration in constructed wetlands (CWs). A pilot-scale CW (6 × 600 L) was developed to jointly rhizofiltrate such BME, produce Cu-rich root mat for ecocatalysis and provide usable shoot biomass with low Cu concentration. Three CW units were unplanted control (Ctrl) while three others were planted with Arundo donax L. ( Ad ) in floating racks. The rhizofiltration was carried out during 30 days in the early growing season. Total Cu concentration in the BME was 4.4 mg Cu L −1 . Copper removal peaked within the 48 first hours after Bordeaux mixture addition in the Ad and Ctrl units (i.e. 92 and 81% respectively). The BME Cu concentration met the requirement for indirect discharge of chemical industry effluents (i.e. 0.5 mg Cu L −1 ) at T 48h (0.4 ± 0.2) and T 21days (0.4 ± 0.1) for the Ad and Ctrl units, respectively. At day 30, in the Ad units, Cu concentration remaining in the water and distributed between A. donax roots, shoots was respectively 3.5, 33 and 0.5% of the initial Cu input. In the Ctrl units, Cu remaining in water was low (7%) and Cu removal (93%) could be partly explained by its immobilization in the Cu-rich biofilm (i.e. 207210 ± 18516 mg Cu kg −1 ) coating the vat wall. Foliar chlorophyll (i.e. a, b and total) and carotenoid contents decreased at day 30 but root and shoot dry weight (DW) yields increased by 23% and 47% per Ad unit, respectively. The shoot Cu concentration remained in the common range (i.e. 3–20 mg Cu kg −1 ) while the root Cu concentration reached 623 ± 140 mg kg −1 allowing 786 mg Cu removal by the root mat. Higher Cu concentration in BME or subsequent repetitions of treatment cycle must be tested to achieve at least 1000 mg Cu kg −1 DW in roots (threshold value for Cu-ecocatalyst) whereas the biofilm role must deserve more attention.