Microcystin-LR removal by powdered activated carbon: the influence of natural organic matter in non-bloom and bloom water

Abstract

This bench-scale study examined the impact of natural organic matter (NOM) on microcystin-LR (MC-LR) adsorption to powdered activated carbon (PAC) prior to and during a cyanobacterial bloom by quantifying NOM fractions including humic substances (HS) and biopolymers with liquid chromatography-organic carbon detection (LC-OCD). Three commercially available powdered activated carbons (wood-, coal-, and coconut-based PACs) were applied to the same reservoir water sampled prior to and during a bloom. Under bloom conditions, phytoplankton content was much higher, mainly cyanobacteria, the dissolved organic carbon increased by 57.4%, and the NOM composition changed. The content of biopolymers increased substantially, doubling compared to non-bloom conditions. The change in the amount of other NOM components, including HS, was less notable. For the cyanobacterial bloom water, MC-LR adsorption kinetics slowed for all PACs, while capacities decreased for the coal-based PAC at all contact times, and for the wood-based PAC only at 0.5 h contact time. The microporous coconut-based PAC performed poorly under all conditions studied. HS appear to be in direct competition with MC-LR for adsorption sites in the mesopore volume. In bloom water, the mass of HS adsorbed per unit mesopore volume was lower for wood-based PAC (45 mgC cm⁻³) than for coal-based PAC (106 mgC cm⁻³), indicating that more adsorption sites were available for MC-LR on the wood-based PAC, which was the least impacted by the bloom. Overall, this study provides new insights into the impact of different NOM fractions on MC-LR adsorption by PACs when treating surface water during a toxic cyanobacterial bloom.