Green walls have long been touted as elegant nature-based solutions for treating greywater at its source. They consume little energy, occupy a vertical footprint that fits neatly onto urban buildings, and offer co-benefits ranging from biodiversity support to thermal regulation. But their Achilles heel has always been consistency. Previous studies have shown that green walls struggle to remove phosphorus, can leach nitrate back into the environment, and falter under operational stresses such as drying and rewetting cycles—the very conditions that occur when building occupants go on holiday and the system sits idle. Media saturation, insufficient contact time, and suboptimal plant–substrate combinations have repeatedly been blamed for these shortcomings.

The new study tackles these weaknesses head-on by bolting a post-treatment column, or PTC, onto the effluent side of a green wall planted with Carex secta, a New Zealand native sedge prized for its vigorous root systems and nutrient uptake capacity. The PTC is packed in equal volumetric parts with three complementary adsorbents: calcinated kina shell, derived from sea urchin waste and heated to 800 degrees Celsius to enrich it with calcium oxide and calcium hydroxide; a zinc chloride-activated apricot stone biochar, whose zinc impregnation creates an extensive network of micro- and mesopores; and natural zeolite, a crystalline aluminosilicate with a high cation exchange capacity. Each material was chosen for a specific job—the shell targets phosphate through calcium-mediated precipitation, the biochar adsorbs nitrate, surfactants, and residual organics, and the zeolite mops up ammonium through ion exchange.

The numbers are striking. Across the full two-stage system, overall removal efficiencies reached 95 percent for total organic carbon and 96 percent for chemical oxygen demand. The post-treatment columns alone polished away 99.8 percent of residual ammonia, 95 percent of anionic surfactants, and between 25 and 37 percent of UV-absorbing aromatic compounds, while stripping a further 34 percent of phosphate and 64 percent of the nitrate that had leached from the green wall units. Under the lowest hydraulic loading rate of 283 millimeters per day—corresponding to a hydraulic retention time of nearly 17 hours—the PTC achieved complete phosphate removal throughout the entire experimental period, a feat the researchers attribute to the slow stoichiometric consumption of calcium hydroxide under reduced cumulative loading.

Scanning electron microscopy, X-ray diffraction, and Fourier transform infrared spectroscopy revealed exactly how each adsorbent earned its keep. The calcinated kina shell developed rough, flaky, porous surfaces, and after treatment showed clear evidence of hydroxyapatite and dicalcium phosphate dihydrate precipitation—mineral signatures of phosphate being locked into solid calcium phases. A strong phosphorus-oxygen bond band at 1,043 inverse centimeters in the FTIR spectra confirmed the capture. Notably, unlike in earlier batch experiments where loosely bound calcium flakes washed away as suspended precipitates, the column configuration retained these precipitates between particles, an important practical advantage. The biochar, meanwhile, showed no significant crystallographic changes after treatment, indicating that contaminants were adsorbed within its porous matrix through pore filling, electrostatic interactions, hydrogen bonding, and pi-pi stacking rather than surface complexation. The zeolite’s framework remained structurally intact, with new spectral peaks confirming ammonium adsorption and possible capture of aromatic organics.

The real stress test came when the researchers subjected the system to a two-week drying period followed by greywater with double the pollutant concentrations—a scenario simulating an empty building over a holiday followed by a sudden return to full occupancy. Here the difference between planted and unplanted columns became dramatic. The vegetated green wall saw ammonia rise modestly from 0.02 to 1.63 milligrams per liter and phosphate from 3.42 to 4.44 milligrams per liter, and it rapidly regained full treatment efficiency once flows resumed. The non-planted control, by contrast, deteriorated severely: ammonia surged nearly tenfold to 8.88 milligrams per liter and phosphate doubled to 8.32 milligrams per liter, and after recovery the control continued releasing ammonia at ten times and phosphate at twice the rates of the planted system. The PTCs, remarkably, maintained their pollutant removal throughout the stress period, with the adsorbent matrix buffering against both hydraulic and pollutant shocks.

Metagenomic sequencing of the 16S rRNA gene shed light on why vegetation matters so much. The planted columns hosted a richer and more even microbial community, enriched with rhizosphere-associated genera such as Bradyrhizobium, Burkholderia, Dyella, and Opitutus—taxa known for nitrogen fixation, phosphate solubilization, and carbon degradation. The unplanted columns were dominated by a narrower assemblage of aerobic oligotrophs like Pseudomonas, Mycobacterium, and Cupriavidus. Intriguingly, the researchers found no clear dominance of classical nitrifying genera such as Nitrosomonas or Nitrospira, suggesting that nitrogen transformations were carried out by low-abundance specialists or distributed across multiple taxa, possibly including heterotrophic nitrification coupled with aerobic denitrification—a process that can operate under the fluctuating oxygen conditions that root oxygen loss creates in the rhizosphere.

The plant itself proved remarkably resilient. Carex secta shoots, cut flush with the media surface at the start of the experimental period, regrew to 27 centimeters within a month and 55 centimeters by the second month, confirming the species’ suitability for long-term use in living treatment systems. Its radial oxygen loss—the release of oxygen from roots into the surrounding media—kept dissolved oxygen levels significantly higher in the planted column, supporting the aerobic microbial activity that drove near-complete ammonia removal and contributed to the system’s redox buffering capacity under stress.

From a practical standpoint, the economics look feasible. Based on phosphate breakthrough under the lowest loading condition, the adsorbent mixture is projected to last roughly 70 to 75 days before exhaustion, corresponding to a quarterly replacement interval that could be extended by deepening the column or regenerating the media. Scaling to typical household greywater production of 100 to 150 liters per person per day, the researchers estimate a requirement of only 2 to 3 kilograms of the mixed adsorbents per person per month. Zeolite costs around 75 to 350 New Zealand dollars per tonne in bulk, while biochar runs roughly 2 to 4 dollars per kilogram—modest sums compared with green wall installation costs of 640 to 2,300 New Zealand dollars per square meter.

Caveats remain. Pathogens were not assessed in this study, so any reuse-oriented deployment would still require a disinfection barrier such as ultraviolet treatment. The zinc chloride activation of the biochar raises questions about potential zinc leaching under varying pH and ionic strength, which the authors flag as a priority for targeted safety testing. Long-term exhaustion rates under seasonal conditions, regeneration and beneficial reuse of spent media as soil amendments, and validation with real rather than synthetic greywater under outdoor climates all await further study. Yet the core message stands: by combining the biological elegance of a planted green wall with the targeted chemistry of waste-derived adsorbents, this two-stage design offers a scalable, low-maintenance pathway toward treating greywater where it is produced—turning a household waste stream into a resource, one sea urchin shell and apricot pit at a time.

Subject of Research: Two-stage green wall greywater treatment using waste-derived adsorbents

Article Title: Enhanced greywater treatment in a two-stage green wall system by incorporating waste-derived adsorbents

Article References: Gholami, M., O’Sullivan, A. D., Tan, Y., Hao, T., & Mackey, H. R. (2026). Enhanced greywater treatment in a two-stage green wall system by incorporating waste-derived adsorbents. Frontiers in Environmental Engineering, 5, Article 1924831. https://doi.org/10.3389/fenve.2026.1924831

Image Credits: AI Generated

DOI: 10.3389/fenve.2026.1924831

Keywords: greywater, green walls, waste-derived adsorbents, biochar, zeolite, phosphate removal, nitrogen removal, nature-based solutions, water reuse, Carex secta, decentralized treatment, circular economy