Polyethylene (PE) accounts for about 60% of global plastic waste and persists in soils because of its chemical inertness. Conventional remediation, including incineration and landfilling, is energy-intensive and can create secondary pollution. Plant-based approaches are low-cost and ecologically compatible, but quantitative PE degradation remains slow: wheat-soil systems lost only 2.8% over 100 days, and soybean rhizospheres degraded about 8% of poly(butylene adipate-co-terephthalate) (PBAT) microplastics over 70 days.
Carbon nanomaterials (NMs) can accelerate pollutant breakdown in simplified laboratory systems, yet their performance in complex soil-plant systems, over time, and across soil depths is poorly understood. Given these challenges, deeper research is needed into how carbon NMs behave in planted soils and whether they can safely enhance PE rhizoremediation.
Researchers at University College London (UCL), Department of Civil, Environmental and Geomatic Engineering, conducted the study, published (DOI: 10.1016/j.ese.2026.100772) online on 23 September 2026 in Environmental Science and Ecotechnology. The team used a 90-day alfalfa-soil experiment to test MWCNTs and graphene oxide (GO) for PE degradation, tracking PE mass loss, surface chemistry, mechanical strength, plant growth, and microbial community changes.
In unamended planted controls, PE loss remained low: 0.6% at the surface and 1.2% at the bottom after 90 days. Without plants, PE plus NMs showed no detectable degradation. MWCNTs produced the strongest surface effect, reaching 12.1% loss at 200 mg kg-1 after 90 days and 9.1% by day 30, while bottom-layer loss plateaued near 6%. GO was weaker overall but most effective at 150 mg kg-1, reaching 7.0% at the surface and 5.8% in the bottom layer by 90 days.
Mechanical testing showed pristine PE at 782.8 ± 0.3 MPa; planted treatments reduced modulus to 23.2 MPa at the bottom and 19.6 MPa at the surface, with multi-walled carbon nanotubes (MWCNTs) causing the greatest loss. Fourier-transform infrared (FT-IR) and X-ray photoelectron spectroscopy (XPS) indicated surface oxidation, with an O 1s/C 1s ratio of 0.49 in surface films from the plant-MWCNT treatment versus 0.17 for pristine PE. MWCNTs enriched early degraders such as Pseudolabrys and Oleiharenicola and predicted Kyoto Encyclopedia of Genes and Genomes (KEGG) ortholog K00799, whereas GO sustained Rhodanobacter and Gemmatimonas and K02003. PE alone reduced germination potential by 32–35% and germination rate by 16–18%; NMs alleviated this phytotoxicity.
The authors said the key advance was separating rapid surface chemistry from slower root-driven processes. They said MWCNTs appeared to act almost like an early catalyst at the surface, while GO supported a more sustained microbial response deeper in the soil. They also said the plant was essential: without alfalfa, no measurable PE degradation occurred even with nanomaterials. They added that the results are a proof of concept, not a field-ready recipe, because dose, soil type, nanomaterial fate, and non-target effects still need testing.
The findings point to two possible applications. MWCNTs may suit short-term, intensive cleanup of surface-contaminated hotspots, where rapid oxidation and mass loss are priorities. GO may be better explored for longer, deeper soil restoration in synergy with plant roots, especially at optimized doses. But the study also found lower insect-trap counts in most nanomaterial treatments, signaling possible non-target effects. Dose optimization, worker-safety controls, cost reductions, and life-cycle assessment are needed before field use. Commercial PE films also contain additives such as carbon black, whose interactions with nanomaterials and degradation intermediates require further study. Together, these findings highlight the need to balance degradation efficiency with ecological safety.