Water hyacinth–derived biofilters for industrial air pollution mitigation and circular blue economy applications

Water hyacinth is an invasive freshwater macrophyte whose aerenchymatous stem tissue has been proposed as a biodegradable replacement for synthetic air filter media. Performance figures reported for such filters normally rest on two shortcuts: the medium is treated as a uniform mat of fibres, and removal efficiency is read off a pair of low cost gas sensors placed upstream and downstream. This work shows that both are unsafe. Aerenchyma is bimodal: open channels of the order of one hundred micrometres coexist with dense lignocellulosic walls, and the two act as parallel hydraulic paths. A dual-scale model is derived in which flow divides by the permeability of each phase, and at representative morphology about half the air bypasses the collecting fibres. The uniform description reproduces size-resolved efficiency to within roughly one percentage point, yet underestimates pressure drop by a factor of 2.4, so that the quality factor inferred from it is too high by a factor of 2.3 to 3.6. Under dust loading the walls block faster than the channels, the bypass fraction grows, and fine particle efficiency falls from 59.0 to 46.6 percent over ninety days while pressure drop rises by only 54 percent. A conventional filter behaves in the opposite sense, so a pressure-triggered replacement rule fires about four months too late. Paired metal oxide sensors are shown to amplify any multiplicative instrument error by a factor of about 2.4, so that a twenty percent mismatch in baseline resistance combined with humidity cross-sensitivity turns a true removal of eighteen percent into an apparent fifty four percent; a bypass cross-calibration protocol recovers the true value to within 2.3 percentage points. Machine learning on the resulting stream predicts remaining service life to about twenty three days against a mean life of sixty two days, and classifies saturation state at an accuracy of 0.68, well below values commonly quoted. A life cycle balance gives a net saving of seventy one kilograms of carbon dioxide equivalent per tonne of wet biomass, but the saving reverses if more than thirteen percent of the water is removed thermally. All results reported here are obtained by simulation.