Nickel agromining has, until now, been almost synonymous with one landscape: natural ultramafic (serpentine) ground, cropped with an elite Mediterranean hyperaccumulator such as Odontarrhena chalcidica. A new open-access study from the University of Turin, Giunchino and colleagues (2026) in the Journal of Hazardous Materials Advances, asks a different question: does the whole value chain — take up the nickel, harvest and ash the biomass, then recover metal from the ash — hold together on anthropogenic urban soil, using plants you might actually find on a contaminated city lot? It is one pot experiment, not a field, and its answer is a useful mix of encouragement and caution.
What the trial did
The authors grew three functionally different species — the Brassicaceae hyperaccumulator Alyssum argenteum, the ruderal Artemisia vulgaris (mugwort), and the forage grass Dactylis glomerata (cocksfoot) — in an urban soil and a nickel-enriched variant of it, comparing phytoextraction against phytostabilization behaviour. Crucially, they did not stop at the plant: harvested biomass was ashed and the nickel recovered by a hydrometallurgical extraction followed by gravimetric (precipitation–redissolution) separation. That is the agromining loop in miniature — bio-ore to metal — attempted on a soil that never saw serpentine.
What clears the bar, and what that is worth
A. argenteum behaved as a genuine hyperaccumulator: foliar nickel above 1,000 mg kg⁻¹ dry weight — that is, above the canonical 1,000 µg g⁻¹ threshold for nickel, the two units being identical — with a translocation factor greater than 1, so nickel moved preferentially to the shoots, the trait agromining depends on. The abstract gives no exact figure, only that the shoots cleared the line, so we quote none.
Two things temper the enthusiasm. First, note what is not new here: A. argenteum is an already-known serpentine nickel hyperaccumulator, so clearing the threshold is not a discovery — and it was cleared in nickel-enriched pot soil rather than on natural substrate, a point strict definitions of hyperaccumulation care about. The genuinely new element is the setting: the clade’s trait, and the recovery chemistry, both surviving the move to urban made ground. Second, “>1,000 µg g⁻¹” clears the bar but sits far below the plants the industry actually uses. Our database records the agromining reference crop O. chalcidica at foliar nickel up to ~22,000 µg g⁻¹ (about 2.2%), and the benchmark laboratory hyperaccumulator Alyssum lesbiacum to ~20,000 µg g⁻¹ — both literature maxima. So A. argenteum here is at least ~20-fold below those ceilings on concentration alone, and the gap in recoverable nickel per hectare is likely larger still, since yield is biomass × concentration and a small Alyssum is unlikely to match the high biomass that helps make O. chalcidica a crop. The news is not a better crop; it is proof that an integrated grow–ash–recover chain can run on urban feedstock.
The other two species mark the approach’s boundaries. D. glomerata took nickel up mainly into its roots with limited translocation — a root-sink, phytostabilization profile rather than an extraction one — and, while the authors note it tolerated and accumulated nickel better than the mugwort, it still showed visible chlorosis and significantly less biomass than controls: nickel it could hold, but not without a toxicity cost. Note the corollary, because it runs against intuition: precisely because the nickel stays in the roots, the grazed shoots carry comparatively little, so root-dominant partitioning argues against a straightforward forage-to-grazer transfer, not for it — the real red flag for D. glomerata is phytotoxicity, not a leaf that poisons a cow. A. vulgaris showed poor resilience and low accumulation. Three species, three outcomes, one soil.
The result to sit with: the labile pool shrinks
The finding most worth dwelling on is easy to skim. Over the experiment, the readily mobile nickel fraction the authors measured early on decreased with time. The authors read this as progressive fixation of nickel within the soil matrix — and that reading is theirs, an interpretation, not something the abstract-level data we could verify prove on their own. It is worth being explicit about the ambiguity: a falling labile pool is also exactly what plant uptake produces, and without the study’s full mass balance (in results tables we could not access) one cannot cleanly separate soil re-sequestration from removal by roots. The rhizosphere of a hyperaccumulator, moreover, actively mobilises nickel, so a declining bulk labile pool need not mean the plant itself was starved of metal.
Alongside this, principal-component analysis associated the nickel-enriched soil with higher ammonium, nitrite and sulfate and with raised ionic strength, and the elemental data showed reduced uptake of several divalent elements — a pattern consistent with nickel competing with other divalent cations at the root, though PCA shows co-variation, not a proven mechanism, and some of the “nickel” signal may travel with the enriching salt and its counter-ion rather than with nickel as such.
The defensible take-home for the field
Strip the interpretation back to what is firm and one lesson stands out, and it is a caution about extrapolation. A freshly nickel-spiked soil starts with an unusually mobile metal pool; metal “ageing” — the slow drift of added metal into less-available forms — is textbook soil chemistry, and it means a spiked pot will tend to overstate what aged, real-world contamination gives up to a crop. Whether or not the decline in this trial is fixation or uptake, the broader point holds: pot phytoextraction rates should not be read straight across to a field remediation timetable. Where it is fixation, the practical implication is that extraction can be limited by declining bioavailability however willing the plant, which argues for mobilising amendments or repeated cropping and for tracking the labile pool through the season rather than measuring only the tissue at the end. That is a hypothesis this one pot trial raises, not one it settles.
Closing the loop — with a bound on the claim
The paper’s other contribution is that it actually recovered nickel from the ashed biomass rather than only measuring it in leaves — the step that separates phytomining from a plant survey, here demonstrated end-to-end at proof-of-concept on urban feedstock. We flag a limit on our own reading, in keeping with this site’s rule: we verified the study’s abstract, highlights and its table and figure captions directly from the publisher, but the full results tables sit behind a bot-block we could not pass, so we do not quote a recovery efficiency, an ash yield, or any exact soil or tissue nickel value — those live in tables we could not read, and we will not restate numbers we did not see.
What to take from it
One pot trial has shown that an Alyssum-clade hyperaccumulator clears the nickel bar on urban soil and that the harvested nickel can be recovered from its ash — the agromining chain working off-serpentine, at bench scale. It has also shown the catch: the soil’s readily mobile nickel fell over the run, a reminder that a pot’s early bioavailability is not the field’s steady state. The useful next steps are unglamorous: replicate on aged field contamination rather than a fresh spike, resolve uptake versus fixation with a full mass balance, track the labile pool through the season, and report the recovery numbers in full.
Provenance: this analysis is grounded in the primary source — Giunchino F, Mucciarelli M, Angus FL, Sordello F, Borrelli S, Lanfranco L & Calza P (2026), “Nickel uptake, partitioning, and recovery in three plant species grown in an enriched urban soil,” Journal of Hazardous Materials Advances 23:101392, doi:10.1016/j.hazadv.2026.101392, gold open access (CC BY-NC-ND), University of Turin. We retrieved and read the article’s abstract, highlights and table/figure captions from the publisher but could not access the full results tables (the open-access PDF is served behind a Cloudflare challenge that blocked every retrieval route we tried); every quantitative claim above is bounded to what those verified sections state — “foliar nickel >1,000 mg kg⁻¹ dry weight,” “translocation factor >1,” the declining mobile-nickel fraction, and the divalent-cation association — and no table-level value is restated. Those load-bearing statements, and the two context figures (the 1,000 µg g⁻¹ nickel threshold of van der Ent et al. 2013; the mg kg⁻¹ ≡ µg g⁻¹ unit identity), were cross-checked against the source across two independent model vendors (Claude and GPT), each re-retrieving the primary source separately, plus an independent web re-retrieval; both confirmed every claim verbatim. Our intended third-vendor coder reviewers (Kimi and GLM) and a Gemini fallback were all offline on the host this run, so the usual three-vendor check ran two-deep — noted here for transparency. Comparison figures for Odontarrhena chalcidica and Alyssum lesbiacum follow the compilation in our database; see our methodology for how records are verified.