Biological microplastic removal in wastewater

Municipal wastewater plants retain, across most published mass-balance studies, between 90 and well over 99 percent of the microplastic particles arriving in the influent. The figure is real and is almost always read wrongly: it describes where the particles went, not what happened to them.

The capture mechanism

No stage of the conventional train was designed for polymer particles, and yet two of them work. Primary settling splits the stream by density: polyethylene and polypropylene at roughly 0.90–0.95 g/cm³ float and are skimmed with the scum, while polyethylene terephthalate at 1.38 and PVC near 1.4 settle. Polymers of intermediate density, and fibres whose behaviour is governed by drag rather than density, carry through.

The biological stage does the bulk of the work, and it does it by flocculation. Bacteria in activated-sludge flocs secrete extracellular polymeric substances — proteins and polysaccharides — that bridge particles into aggregates. A polymer particle is incorporated into that aggregate like any other inert mineral nucleus. In parallel, a biofilm develops on its surface within days; the mass of that film and of the mineral matter it traps raises the particle’s effective density above that of water, and material that arrived buoyant begins to settle. This is what “biological filtration” of microplastics consists of — heteroaggregation and biofouling, not any enzyme acting on the polymer.

Why this is transfer, not removal

The retained microplastic ends up in the waste sludge. Sludge is dewatered and, in a large fraction of cases, applied to land, which closes the mass loop: the particles return to soil, where polyolefins do not degrade (see microplastic biodegradation on land). Directive 86/278/EEC on sewage sludge in agriculture regulates heavy metals and pathogens and is silent on polymers; the recast urban wastewater directive, (EU) 2024/3019, introduces monitoring of microplastics in influent, effluent and sludge — that is, it requires measurement rather than removal.

The measurement floor

A reported removal percentage cannot be separated from the mesh that sampled it. Most mass balances use nets and filters with apertures between 20 and 300 micrometres, and “99 percent” refers to what that aperture retained. Particles of a few micrometres, and the nanoscale fraction, appear in neither the numerator nor the denominator. Meanwhile the shear of mixing, aeration and pumping inside the works itself shifts the size distribution downward.

A membrane bioreactor removes part of that uncertainty — ultrafiltration pore sizes are nominally 0.04–0.4 micrometres — but it too relies mainly on the cake layer, and an elongated fibre can thread an aperture once aligned with the flow. Its central trade-off sits in the same place: the extracellular polymers that capture particles into flocs form a gel layer on the membrane and drive transmembrane pressure up. The removal mechanism and the fouling mechanism are the same one.

Degradation inside the works is not on the table at all: hydraulic residence time is hours, while hydrolysis of even the most vulnerable polyester runs in months and years.