The method at the center of the debate is the material intensity approach, the workhorse of bottom-up building material stock analysis. It works by multiplying a building’s gross floor area by an aggregated coefficient, expressed in kilograms or cubic meters of material per square meter of floor space. That simplicity is precisely why it dominates the field: it requires no component-level data, no floor plans, and no knowledge of internal layouts, making it feasible to estimate material stocks for entire cities, nations, or the global building stock. Those estimates matter enormously for the circular economy, because knowing what materials sit where in the built environment is the prerequisite for urban mining, the strategic recovery of steel, concrete, and timber from buildings at the end of their lives.

The problem, as Yingqi Jia and Chen Feng point out, is that the material intensity method treats all buildings of a given type as essentially interchangeable. Yet real buildings of the same class, built in the same era, can differ dramatically in footprint shape, corridor arrangement, number of units, room layouts, number of floors, and structural system configuration. Quantifying how much these design choices matter has long been blocked by a practical obstacle: detailed internal layout data for real buildings is scarce, and no city keeps a comprehensive database of floor plans suitable for this kind of controlled comparison. The researchers’ solution was to sidestep real-world data entirely and generate their own.

Their pipeline is a hybrid of two computational approaches, each compensating for the other’s weaknesses. Parametric modeling offers precise, systematic control over design parameters, which is exactly what a controlled experiment requires, but achieving architectural realism with parameters alone demands an impractically complex rule set. Deep learning generative models, by contrast, can produce realistic floorplans with minimal manual effort, but they do not allow explicit, systematic control of the variables under study. The team’s workflow combines both: parametric routines generate footprints, insert corridors, and subdivide space into units using a corridor-aware binary space partitioning algorithm, while a neural network called Graph2Plan, pre-trained on real-world floorplans, generates plausible room layouts within each unit. The pipeline then assembles three-dimensional structural components, resolves intersecting volumes with Boolean operations to avoid double counting, and calculates the material volume of every wall, column, beam, and slab.

Running this pipeline across the full parameter space produced 48,600 synthetic residential buildings, all with a constant gross floor area of 2,400 square meters, while systematically varying footprint shape among square, rectangle, L-shape, T-shape, and H-shape geometries; corridor type among single-loaded, double-loaded, and point-access systems; unit counts; room layouts; floor counts from one to six; and structural grid spacing and member sizing. The headline result is striking: total material stock ranged from 758 to 2,888 cubic meters across this design space, with a mean of 1,548 cubic meters and a standard deviation of 320. Under the traditional material intensity method, every one of those buildings would have been assigned the identical stock value, because floor area alone determines the estimate.

Disentangling which design parameters drive this variability required statistical analysis rather than inspection of individual cases. A multiple regression model explained more than 95 percent of the variance in material stock and revealed a clear hierarchy of influence. The number of floors was the single most powerful factor, alone accounting for roughly one-third of the variation, followed by the sizing of the structural frame, then the number of units and the footprint shape. Grid spacing and floor height came next, while corridor type and room layout had much smaller, though still statistically significant, effects. Every parameter in the model was significant at the p < 0.001 level.

The physical mechanisms behind these rankings are intuitive once spelled out. Holding floor area constant, the number of floors controls a building’s slenderness, and every additional floor must be enclosed by walls. In the synthetic dataset, wall volume grew from 600 cubic meters in single-story buildings to 1,186 cubic meters in six-story buildings, driving a 55.2 percent increase in total stock, while the structural frame grew far more modestly. Structural sizing mattered even more sharply within the frame itself: moving from slim to bulky member dimensions raised total material volume by 35.1 percent, and widening the structural grid from 4.0 to 8.0 meters cut it by 10.5 percent by reducing the number of columns and beams needed. Footprint shape operated through the surface-to-volume ratio, with the sprawling H-shape demanding 12.1 percent more material than the compact square, almost entirely in wall components, since frame volumes stayed nearly constant across shapes. Adding units raised stock sublinearly, because each new partition needs proportionally less wall, and going from one to three bedrooms per unit added only 5.4 percent.

Because the dataset is synthetic, the authors took plausibility seriously. They reproduced wall and slab volumes from an open residential building information model within 4 percent, checked that the 2,400 square meter floor area falls within real-world ranges observed in OpenStreetMap records across 17 cities on six continents, and compared converted mass-based intensities against the RASMI global benchmark database, producing a median of 1,276.77 kilograms per square meter against a benchmark of 1,285.52. They also examined structural calculation records for two real Hong Kong apartment towers, Tin Sam Villa and Millennium Court, finding derived column-and-beam volumes of 225.1 and 226.2 cubic meters, both close to the synthetic mean of 222.2 and well inside the synthetic range. Supplementary tests at 1,200 and 4,800 square meters confirmed the findings are robust to the choice of fixed floor area.

The practical implications cut in two directions. For analysts, the results suggest that material intensity databases should expand their building typologies beyond use type and construction period to include footprint shape, number of floors, and structural configuration, the parameters that dominate design-induced variability. Where such classification is not feasible, the pipeline can generate exploratory adjustment factors, nudging baseline intensities upward for slender, complex buildings and downward for compact, low-rise ones. For designers, the study offers a menu of material-efficiency levers: reduce slenderness, widen structural grids, slim down frame members, minimize unit divisions and floor height, favor square or rectangular footprints, and use point-access cores instead of corridors. The authors caution that these levers must be weighed against structural, regulatory, economic, and functional requirements they did not model, and that their analysis was limited to reinforced concrete frames, excluded foundations, roofs, and mechanical systems, and explored a theoretical design space rather than the distribution of real buildings, so the reported range represents potential variability rather than expected real-world spread.

What may prove most durable about the work is the methodological framework itself. By fusing parametric modeling, generative AI, and component-based material accounting into a fully automated pipeline, the researchers have created an extensible tool for running what-if experiments on building design and resource demand at a scale impossible with real-world data. The complete workflow, the database of more than 48,000 building configurations, and tutorials for reproducing the results are openly available on GitHub. As cities race to turn their building stocks into material banks for a circular economy, the study delivers a clear warning and a clear opportunity: floor area alone cannot tell you what a building is made of, but the design decisions that shape it can now be measured, ranked, and, ultimately, engineered for less.

Subject of Research: Sensitivity of building material stock estimates to architectural design parameters

Article Title: Assessing the sensitivity of material-intensity-based building stock estimates to design parameters

Article References: Jia, Y., & Feng, C. (2026). Assessing the sensitivity of material-intensity-based building stock estimates to design parameters. Journal of Industrial Ecology, 30(4), 2035-2050. https://doi.org/10.1007/s44498-026-00138-5

Image Credits: AI Generated

DOI: 10.1007/s44498-026-00138-5

Keywords: building material stock, material intensity, circular economy, urban mining, parametric modeling, deep learning, floorplan generation, structural design, reinforced concrete, industrial ecology, synthetic data, material efficiency

The research, led by Valentina Pusateri of the Technical University of Denmark together with Mikołaj Owsianiak, Marja Rinne, Stig I. Olsen, Michael Z. Hauschild, and Sami Kara, tackles a stubborn blind spot in environmental accounting. Life cycle assessment, the standard tool for measuring the environmental footprint of products, has long struggled with how to treat metal resources. Recent methodological advances have shifted the focus from simple resource depletion toward resource dissipation: the idea that a metal becomes a genuine loss when it flows into a sink from which future users cannot realistically recover it. But existing methods typically treat all unrecovered metals as equally dissipated, ignoring the crucial fact that some stocks are far easier to tap than others, and that accessibility shifts as technologies mature and ore grades decline.

The Danish-Australian team proposed an elegantly simple criterion built on comparative profitability. A metal in a given anthropogenic stock is classified as dissipated when the net present value of extracting it from that stock is lower than the net present value of extracting the same metal from the reference stock, which the researchers defined as the upper continental crust, the most plentiful and dominant source of metals. In other words, if mining companies can always make more profit pulling copper out of the ground than out of a landfill, the landfilled copper is effectively inaccessible, no matter how concentrated it is. The authors were careful to stress that this does not mean the metal cannot physically be extracted; it means that in a global economy with a plentiful alternative source, the stock will not become the dominant supply, and its metal functions remain out of reach for the system as a whole.

To operationalize this dissipation quotient, the team assembled an unusually comprehensive economic dataset. They screened more than 500 publications and reports, ultimately extracting capital and operating expenditure data from 45 studies covering extraction of gold, copper, and iron from mine tailings, landfills, and hoarded stock such as end-of-life electronics. All costs were harmonized to 2024 euros using GDP deflators and exchange rates, and allocated per kilogram of extracted metal. Because revenues from metals are set on global exchange markets regardless of origin, the comparison reduces to a battle of costs: whichever source delivers metal more cheaply wins, and the loser is classified as dissipated.

The numbers reveal stark differences between the three metals. Extracting copper from anthropogenic stocks costs between roughly 6 and 670 euros per kilogram, while gold extraction runs between 28,000 and 44,000 euros per kilogram, a gap of about three orders of magnitude driven by the vastly different concentrations of the metals in waste streams. Iron sits at the bottom of the range, from about 0.07 to 120 euros per kilogram. Hoarded stock, meaning discarded phones, computers, and circuit boards, proved the most expensive source overall, despite its high metal concentrations, because collecting and dismantling small, complex devices is costly. Landfills showed the lowest average extraction costs for copper and iron, while tailings generally emerged as the most accessible stock type across the analysis.

The crucial twist comes from the reference side of the equation. Ore grades have fallen steadily since large-scale mining began, and the researchers projected this decline forward using an exponential decline model calibrated on historical data, cross-checked with a power regression approach. Copper ore grades have historically declined by about 1.72 percent per year, gold by about 1.05 percent, and iron by 0.61 percent. Because energy for mining and processing is the dominant cost driver and rises steeply as grades fall, the cost of primary extraction is expected to climb. At a 25-year horizon, the projected grade decline raises reference extraction costs by less than 10 percent. But by 300 years, the present value of costs is expected to increase by roughly 20 percent for gold, 200 percent for copper, and a striking 330 percent for iron relative to today.

Running the dissipation criterion across three time horizons of 25, 100, and 300 years produced a nuanced and somewhat counterintuitive picture. Accessibility generally increases with time, as primary mining becomes more expensive and the economic gap narrows. Gold consistently emerged as the most accessible metal, classified as non-dissipated in hoarded electronic waste, which aligns with the fact that gold recovery is already the main economic driver of e-waste recycling. Iron in tailings also performed relatively well. By contrast, hoarded iron, drawn from electronic waste where iron is merely a structural contaminant rather than a recovery target, was classified as dissipated across the entire range of uncertainty at every time horizon, with cost differences reaching minus 311 euros per kilogram at 300 years. Most other metal-stock combinations remained, on average, dissipated even at 300 years, though the uncertainty ranges were wide enough that many straddled the boundary.

The team then asked whether technological learning could flip these classifications. Using the classic Wright’s law framework, which links cost reduction to growth in cumulative output, they calculated the learning rates that emerging extraction technologies would need to break even with primary mining. For most combinations the required learning rates ranged from about 40 to 80 percent per doubling of output, far above the 15 to 25 percent considered realistic for emerging technologies, making a reclassification unlikely. Tailings were the notable exception. Copper recovery from tailings would need learning rates of only 25 to 40 percent, gold less than 20 percent, values well within reach as technologies mature. This finding dovetails with real-world developments, as copper is already being extracted from old mine tailings that were once written off as inaccessible.

The authors are candid about the limitations. Data availability was uneven, with no cost data at all for gold in landfills, and abandoned or dispersed technosphere stocks had to be excluded entirely. The assumption that anthropogenic metal concentrations stay constant over time may not hold if these stocks become active supply sources, and the assumption that metals recovered from waste fetch the same market price as virgin metal may fail where purity is lower. Yet sensitivity analysis suggested that the classifications are more sensitive to technology maturity and learning than to plausible shifts in reference ore grades. Recognizing this irreducible uncertainty, the researchers recommend a careful vocabulary shift: stocks meeting the criterion should be labeled potentially dissipative, and those clearing it potentially non-dissipative, rather than treated as certainties.

The implications reach well beyond academic methodology. Current footprinting approaches, including influential methods that assume everything not recycled within a given timeframe is dissipated, may systematically overestimate resource use impacts by ignoring the metal-, stock-, and time-specific nature of accessibility. The dissipation curves and breakthrough times generated by this framework could feed directly into next-generation life cycle impact assessment metrics, replacing crude recycling-rate proxies with economically grounded estimates of when, and whether, the metals we bury today will serve future generations. In a world where maximum technical circularity is estimated at only 30 to 40 percent and primary extraction will dominate for decades, knowing which of our discarded riches are truly lost, and which are merely waiting for the economics to turn, is a question whose answer is finally taking quantitative shape.

Subject of Research: Economic dissipation of copper, gold, and iron in anthropogenic stocks for resource footprinting

Article Title: Determining the dissipation of copper, gold, and iron resources in anthropogenic stocks based on extraction economics

Article References: Pusateri, V., Owsianiak, M., Rinne, M., Olsen, S. I., Hauschild, M. Z., & Kara, S. (2026). Determining the dissipation of copper, gold, and iron resources in anthropogenic stocks based on extraction economics. Journal of Industrial Ecology, 30(4), 1743-1759. https://doi.org/10.1007/s44498-026-00118-9

Image Credits: AI Generated

DOI: 10.1007/s44498-026-00118-9

Keywords: anthropogenic stocks, resource dissipation, life cycle assessment, circular economy, mine tailings, landfill mining, urban mining, ore grade decline, extraction economics, metal recycling, resource footprinting, technosphere

The research team, led by Jingwei Hou and Yong Geng, applied dynamic material flow analysis (dMFA), a mass-conservation accounting framework that tracks every tonne of a substance through extraction, refining, manufacturing, use, and disposal. Their system boundary covered mainland China and divided the boron industry chain into seven stages: mining and beneficiation, refining and separation, smelting, fabrication, manufacture, use, and waste management and recycling. Thirty-eight boron-containing commodities were tracked, with trade data drawn from China Customs Yearbooks and the UN Comtrade database, production statistics from the U.S. Geological Survey and the CBC Metal database, and process parameters validated through field visits to major producers and ten face-to-face interviews with industry executives and technical personnel.

The methodological machinery is considerable. Process flows were calculated by multiplying production quantities by boron concentrations, losses were applied at each stage using documented loss rates, and the accumulation and retirement of boron in products was modeled with a Weibull probability distribution, the standard statistical tool for describing how products fail and exit service over time. To test robustness, the team ran a Monte Carlo simulation with 10,000 iterations, assigning coefficients of variation of 5 percent to reliable official statistics and 10 percent to literature-derived parameters. The results showed uncertainty below 5 percent for imports, exports, and consumption, and 13 to 19 percent for end-of-life flows, narrow enough that the study’s central conclusions hold firm.

The headline numbers are sobering. Over the twenty-year period, net imported boron at the mining and refining stages reached 2,784 kilotonnes, while domestic extraction contributed only 1,026.7 kilotonnes. China’s reserves, ranked fifth globally at 9.1 million tonnes of boron trioxide equivalent, are concentrated in Qinghai and Liaoning provinces but suffer from low ore grades that make domestic mining economically punishing. Turkey, which holds 950 million tonnes of reserves, dominates the global market and supplied roughly half of China’s total boron imports, with borax, boron ores, boric acid, and boron oxides making up 99.5 percent of that bilateral trade. Such concentration in a single supplier exposes China to geopolitical, trade-policy, and natural-disaster risks.

Consumption tells its own story of transformation. Annual boron use climbed from 41.3 kilotonnes in 2004 to a peak of 161.8 kilotonnes in 2021, before easing to 94.6 kilotonnes in 2023 as upstream mining constraints bit. Glass, ceramics, and enamel products consistently absorbed more than 40 percent of total consumption, with the chemical industry, spanning disinfectants, flame retardants, and catalysts, taking another 24 percent. But the fastest growth came from emerging technologies: by 2020, new energy vehicles and batteries accounted for over 4 percent and 8 percent of consumption respectively, a share that seemed set to expand before supply bottlenecks curtailed it.

The losses along the chain are enormous. Refining borax generates roughly four tonnes of boron mud for every tonne of product, and China discharges about two million tonnes of this waste annually, with boron content of only 0.76 percent. Across the full period, total boron losses reached 1,757.1 kilotonnes, while just 24 kilotonnes were recycled, an overall recycling rate of approximately 1.37 percent. Chemical extraction from boron mud requires acid leaching with high reagent and energy costs and risks secondary pollution, so most recovery involves low-value physical processes such as sintering the mud into construction materials. By contrast, Turkey recovers more than 90 percent of boron from industrial wastes, a benchmark the authors argue China should urgently study.

Meanwhile, an invisible reservoir is quietly building. In-use boron stocks, the element embedded in products currently serving society, nearly doubled during the second decade of the study, rising from 502.5 kilotonnes in 2014 to 966.8 kilotonnes in 2023. Glass, ceramics, and enamel dominate this urban mine at 664.2 kilotonnes, followed by batteries, which grew from 25.7 to 68.7 kilotonnes over the decade, earthmoving machinery, glazed tiles, and aircraft. End-of-life flows, delayed by the roughly 16.5-year average lifespan of boron-containing products, entered rapid growth in the 2010s and reached 31 kilotonnes in 2023, signaling a secondary resource stream that current infrastructure is not equipped to capture.

The trade structure reveals a low-value trap. China imported 2,955.2 kilotonnes of boron over the period, dominated by raw and intermediate materials, then exported 1,257.4 kilotonnes of finished products, chiefly glass and ceramics, glazes, and boron steel, to 225 countries and regions, with Indonesia the largest buyer. Fabrication losses ran at roughly 20 percent, and the authors conclude that the whole boron industry chain remains low-value oriented, generating thin profits while imposing heavy environmental pressure. High-end products such as boron carbide, boron nitride, and rare earth borides are still largely manufactured in developed countries, and Chinese enterprises struggle to achieve large-scale production of these materials essential for energy, defense, and nuclear applications.

The policy prescriptions are concrete. The authors call for increased research and development investment and university-industry cooperation to crack the technological bottlenecks in high-end boron products and boron mud valorization, including promising but unconfirmed routes such as converting the waste into nuclear radiation shielding materials or fertilizers. They recommend a national boron information platform, national standards requiring retired boron-containing products to reach designated disassembly stations, and financial support for recycling enterprises. On supply security, they point to diversification through Chile, Russia, Peru, Argentina, Bolivia, and Kazakhstan, countries aligned with China’s Belt and Road frameworks and recent green minerals cooperation agreements. Economic instruments, including differentiated resource taxes tied to ore recovery rates, carbon taxes on energy-intensive operations, emission fees, and green labeling, round out the governance agenda.

What makes this study resonate beyond China is the method itself. dMFA has previously illuminated the metabolism of steel, aluminum, copper, rare earths, and dozens of critical metals, but boron, a nonmetallic critical material with extreme import dependence, had never received this treatment at national scale. As the world’s clean energy transition accelerates demand for every element that hardens steel, toughens glass, and steadies neutrons, the study warns that a country can dominate manufacturing yet remain dangerously exposed at the raw material frontier, and that the urban mines accumulating in its cities, currently leaking 790.7 kilotonnes of boron into the environment as waste, represent both an untapped opportunity and an unaddressed environmental burden.

Subject of Research: Dynamic material flow analysis of boron flows, stocks, trade, and recycling in China from 2004 to 2023

Article Title: Uncovering the features of boron flows and stocks in China

Article References: Hou, J., Geng, Y., Zhong, C., Gao, Z., & Liu, S. (2026). Uncovering the features of boron flows and stocks in China. Environmental Challenges, 25, Article 101675. https://doi.org/10.1016/j.envc.2026.101675

Image Credits: AI Generated

DOI: 10.1016/j.envc.2026.101675

Keywords: boron, material flow analysis, China, critical materials, recycling, supply chain, boron mud, in-use stocks, Turkey, import dependence, urban mining, resource policy

The stakes could hardly be higher. Primary platinum group metal production is geographically precarious, with South Africa’s Bushveld Igneous Complex supplying roughly 88.7 percent of the world’s platinum group metals and Russia’s palladium-rich Norilsk-Talnakh region alone accounting for about 43.8 percent of global palladium output. Life cycle assessments place the carbon footprint of producing a single kilogram of refined metal at between 28 and 42 tonnes of carbon dioxide equivalent, while the classic aqua regia leaching route releases toxic nitrogen oxide and chlorine-bearing gases. As millions of converters are scrapped each year, spent automotive catalyst has become the obvious secondary resource, but the industry has lacked a clean, low-energy route that works on the real, unprocessed solid material rather than pre-dissolved metal salts.

The Nottingham team, led by Christine Paul with Frankie Rawson, Katalin Kovács and Helena I. Gomes, turned to Cupriavidus metallidurans CH34, a bacterium with almost legendary resistance to toxic metals. Its defensive arsenal sits on two megaplasmids carrying metal-specific gene clusters including cop, czc and cnr, allowing it to shrug off millimolar concentrations of nickel, zinc, copper and cadmium. Crucially, the organism is also electrogenic: it forms biofilms on graphite electrodes and generates current densities of 15 to 150 milliamperes per square metre. Previous bioelectrochemical studies of precious metals had only ever worked with dissolved metal ions in model solutions, or with real catalyst material that had first been chemically leached. This new work, published in Cleaner Engineering and Technology, is the first to put an electroactive culture in direct contact with untreated spent catalyst solids.

The experimental design was elegantly simple. Dual-chamber H-type microbial fuel cells were built from borosilicate glass, with carbon felt electrodes separated by a perfluorinated sulfonic acid membrane and a ferricyanide catholyte chosen specifically so that the cathode would never limit the measured response. The anode chambers received untreated spent catalyst powder at loadings of 10,000, 50,000 and 100,000 parts per million, corresponding to pulp densities of 1, 5 and 10 percent, alongside catalyst-free controls. Eight reactor types, each run in biological triplicate, allowed the team to separate purely chemical effects from genuinely biological ones. After 168 hours of operation, inductively coupled plasma mass spectrometry tracked exactly where the platinum, palladium and rhodium had ended up across the liquid, biomass, electrode and residual solid fractions.

The first surprise came from the biomass data. At the lowest loading of 10,000 parts per million, the bacteria actually grew better than in the catalyst-free control, reaching 221.4 micrograms per millilitre of planktonic protein at 96 hours compared with 201.3 in the control, suggesting the moderate metal challenge stimulated rather than suppressed the culture. But at 50,000 and 100,000 parts per million, growth collapsed to just 60 to 75 percent below the control level. Scanning electron microscopy told part of the story: at the higher loadings, aggregated catalyst particles blanketed the carbon felt electrodes, masking the underlying fibre structure and leaving little visible biological material. The electrode-attached population fared better than the free-floating cells, but the overall picture was one of progressive physical and chemical suffocation.

That suffocation had two distinct sources. The first was acidity. Even in completely abiotic reactors, adding catalyst drove the anolyte pH down in a dose-dependent fashion, from 5.95 in the catalyst-free control to 4.87, 4.06 and 3.93 at the three loadings respectively. This chemical acid load occurred entirely without microbial help and, crucially, preceded the decline in bacterial growth, establishing that acidification was a cause rather than a consequence of the biological collapse. The second source was electrical. Open-circuit voltage peaked at an impressive 0.442 volts at 10,000 parts per million, actually exceeding the catalyst-free control, but fell to around 0.12 to 0.15 volts at higher loadings. Internal resistance told the same story, with activation resistance jumping from 142 ohms in the clean biotic reactor to 3688 ohms at low loading and over 7000 ohms at the highest concentrations, while power density plummeted by more than two orders of magnitude across the range.

Cyclic voltammetry added a further layer of insight. The anodic peak current at 10,000 parts per million reached 688.89 microamperes, more than double the 327.65 microamperes of the catalyst-free control, and the Randles-Ševčík slope, a measure of diffusion-coupled electron transfer, was likewise highest at moderate loading. But the weaker linearity of that fit hinted at extra processes beyond simple diffusion control, and at 50,000 and 100,000 parts per million the peak currents fell to 278.25 and 96.24 microamperes respectively, with increasingly non-Nernstian behaviour indicating quasi-reversible to irreversible electron transfer governed by kinetic limitations. In plain terms, the electrochemical conversation between bacteria and electrode was sharpest at moderate catalyst loading and progressively garbled as particles piled up on the surface.

The metal redistribution results were the most striking of all. Rhodium proved to be the star performer: at 10,000 parts per million, the biotic reactors mobilised 17.93 percent of the rhodium input, compared with just 0.87 percent of platinum and 1.20 percent of palladium. Moreover, 10.3 percent of the rhodium ended up associated with the bacterial biomass, a fraction that was below detection in the abiotic controls, pointing to a genuinely selective biological interaction. The explanation likely lies in catalyst chemistry: road-aged three-way catalysts contain rhodium predominantly in oxidised form, whereas platinum sits mostly as metal. Dissolving metallic platinum group metals requires both a high oxidation potential and a complexing ligand such as chloride, neither of which was present in the anolyte, but rhodium that is already oxidised can bypass that demanding first step. The biological contribution, however, was confined to biosorption and accumulation on the biomass rather than true dissolution of the solid matrix, and the authors are careful to note that no oxidation-state analysis confirmed reduction.

Perhaps the most counterintuitive finding was that mobilisation per gram of catalyst was highest at the lowest loading, despite the more favourable pH there. Platinum recovery per unit mass fell from 2.01 hundredths of a milligram per gram at 10,000 parts per million to 0.332 hundredths at 100,000 parts per million, with palladium and rhodium showing the same pattern. This points to accessible surface area, not acidity, as the true bottleneck: at high solids loadings, particles aggregate, the reactive surface shrinks, and the sheer mass of solid available for re-adsorption pulls dissolved species back out of solution. The same pulp-density penalty has been observed in conventional hydrometallurgical processing, where palladium and platinum recovery drops from above 90 percent at 5 to 8 percent solids to below 70 percent at 20 percent.

The honest conclusion is that this system, as it stands, cannot yet compete with industrial recovery: even at the optimum loading, less than 18 percent of rhodium and under 2 percent of platinum and palladium were mobilised. But the study does something arguably more valuable than demonstrating a working process. It maps the operational window, showing that the transition between sustained and suppressed performance lies somewhere between 10,000 and 50,000 parts per million, and it identifies the two mechanisms, chemical acidification and particulate fouling of the electrode, that constrain it. The authors sketch a roadmap of possible fixes: pH control to decouple acidification from surface effects, pre-washing the catalyst to reduce the acid load at source, fed-batch dosing to keep instantaneous loading within the functional window, or a two-stage configuration that separates mild chemical solubilisation from the bioelectrochemical step that the biology performs best, namely selectively capturing mobilised rhodium. Achieving process-relevant yields, they conclude, will require intervention at the solid-liquid interface rather than tinkering with the microbe itself. For a field desperate to break its dependence on South African ore bodies and 1200-degree furnaces, that is a map worth having.

Subject of Research: Bioelectrochemical recovery of platinum group metals from untreated spent automotive catalyst using Cupriavidus metallidurans

Article Title: Bioelectrochemical treatment of untreated spent car catalyst: Effects of solids loading on platinum group metal redistribution and process limitations

Article References: Paul, C., Rawson, F., Kovács, K., & Gomes, H. I. (2026). Bioelectrochemical treatment of untreated spent car catalyst: Effects of solids loading on platinum group metal redistribution and process limitations. Cleaner Engineering and Technology, 34, Article 101322. https://doi.org/10.1016/j.clet.2026.101322

Image Credits: AI Generated

DOI: 10.1016/j.clet.2026.101322

Keywords: platinum group metals, spent car catalyst, microbial fuel cell, Cupriavidus metallidurans, rhodium recovery, bioelectrochemical systems, urban mining, circular economy, biomining, hydrometallurgy, electroactive bacteria, metal recycling

The research, led by George Yandem, Katarzyna Grygoyć and Magdalena Jabłońska-Czapla of the Polish Academy of Sciences together with Joanna Willner and Tomasz Matuła of the Silesian University of Technology, tackles a question that has received surprisingly little systematic attention: how the way you pre-treat a dead panel interacts with the chemistry of the leaching solution to determine which metals dissolve, and how much. Instead of testing one process in isolation, the team ran a full factorial comparison of three pre-treatment routes against four chemically distinct leaching agents, all under identical ultrasonic conditions, using a decommissioned polycrystalline silicon module originally manufactured by ALGATEC Solar AG in Germany.

The three pre-treatments represented the main strategies competing in the recycling world today. Mechanical milling simply ground the module into particles sieved below 0.5 millimetres. Pyrolysis heated the material to 500 degrees Celsius for two hours, burning away the plastic encapsulants that glue the sandwich of glass, silicon and metal together. Acetone washing softened the ethylene–vinyl acetate encapsulant chemically, allowing the glass to be scraped away and the dissolved polymer to be filtered off under vacuum. Each route was then paired with citric acid, oxalic acid, the chelating agent EDTA, or concentrated nitric acid, and leached in an ultrasonic bath at 40 kilohertz and ambient temperature, with a solid-to-liquid ratio of one gram to twenty millilitres over treatment times from ten to sixty minutes.

Before any leaching began, the team characterised exactly what the waste contained and where those elements ended up after grinding. Using microwave digestion and inductively coupled plasma mass spectrometry, they found the material dominated by copper, tin, lead, antimony, chromium, silver and nickel, with lower but environmentally significant concentrations of cobalt, gallium, germanium, indium, molybdenum and tellurium. Crucially, sieving revealed that most of these elements concentrate in the finest dust. Particles smaller than 0.05 millimetres made up only 13.4 to 14.4 percent of the total mass, yet carried the highest concentrations of most elements. That means a recycler could process just the fine fraction, cutting reagent demand, treatment cost and pollutant load dramatically—an insight with immediate practical value, since industrial-scale crushing can push residues toward concentrations comparable to metal ores.

The headline results concern which combination extracts which element. Pyrolysis followed by nitric acid delivered the strongest overall performance, dissolving up to 300 milligrams per litre of silver, 888 milligrams per litre of copper, 256 milligrams per litre of lead and 109 milligrams per litre of tin. Thermodynamic analysis explains why: tin dissolution in nitric acid carries an equilibrium constant of roughly 10 to the power 37, and lead is barely less favourable, so once the polymer barriers are burned away the oxidation reactions proceed almost irreversibly. Silver, by contrast, is the most noble metal in the set, with a small cell potential of just 0.156 volts, yet the ultrasonically assisted nitric acid route still outperformed earlier studies by a factor of nearly six in leached silver concentration—a difference the authors attribute to the sonication and the fine pre-milling.

The greener acids proved unexpectedly powerful for the rarer, geopolitically sensitive elements. Oxalic acid, especially after pyrolysis or acetone treatment, was the best route for antimony and indium, reaching 5.07 and 8.11 milligrams per kilogram of indium respectively—close to the total indium content of the finest digestible fraction. Citric acid excelled at tellurium and germanium, while EDTA with simple milling pulled out cobalt, and acetone treatment followed by oxalic or citric acid preferentially recovered indium, tin and tellurium. The chemistry behind this selectivity is ligand-driven: citrate ions bind metals through tridentate carboxylate and hydroxyl groups, oxalate forms highly stable bis- and tris-oxalate complexes with indium(III) and gallium(III), and antimony(III) oxalate complexes carry a stability constant that makes dissolution spontaneous. In effect, the organic acids act as molecular tweezers, plucking specific ions from the particle matrix while leaving others behind.

Kinetic analysis added a further layer of control. The elements fell neatly into three time groups: silver, arsenic, cobalt, gallium, germanium, molybdenum and nickel peaked within ten minutes; copper, indium, manganese, lead and tin needed thirty; and chromium, antimony, tellurium, thallium and zinc required the full hour. This pattern reflects where each element sits in the material—surface-accessible phases dissolve quickly, while elements buried deeper in the matrix or held by slower ligand-exchange kinetics take longer. Shrinking-core models, the classical framework for hydrometallurgical dissolution, only fit gallium and germanium, indicating that those two elements leach from deep within the particles under chemical reaction control, with germanium showing R-squared values above 83 percent for EDTA and nitric acid. For everything else, multiple mechanisms operate simultaneously, a reminder that ultrasonic cavitation scrambles the tidy assumptions of conventional leaching theory.

Ultrasound itself deserves attention. Acoustic cavitation—the collapse of microscopic bubbles generated by sound waves—strips blockages from mineral surfaces and shatters particles, expanding the reactive area. Previous work has shown ultrasound can shrink waste particles from 30 micrometres to 1 micrometre and cut chemical activation energy from 34.68 to 6.21 kilojoules per mole, producing dramatic jumps in recovery. Here, the ultrasonic bath allowed leaching at ambient temperature in as little as ten minutes, avoiding the heated, hours-long acid treatments typical of the literature. The method’s principal component analysis further confirmed that pre-treatment dominates the overall variance, with pyrolysis samples clustering apart from milled and acetone-washed material, and low-melting-point elements such as cadmium, gallium, vanadium, zinc and thallium responding distinctly to thermal processing.

The industrial context is moving fast. Recent reporting from the International Energy Agency’s Photovoltaic Power Systems Programme shows commercial recyclers such as SOLARCYCLE, Reiling and SPR already running hybrid lines that combine automated deframing, shredding, mechanical sorting and chemical leaching, with capacities between 20,000 and 63,000 tonnes per year. Downstream, dissolved metals are typically converted into sellable products through clarification, selective separation, concentration and final recovery by precipitation, solvent extraction, ion exchange, cementation or electrolysis. The challenge this study addresses is that these hydrometallurgical stages traditionally depend on nitric, sulfuric and hydrofluoric acids—highly corrosive and toxic reagents whose handling and disposal carry their own environmental burden, threatening the very sustainability credentials that solar recycling is meant to deliver.

The authors are candid about the limits of scaling sound waves. Cavitation intensity becomes spatially uneven in large reactors, potentially producing inconsistent leaching, and the energy cost of prolonged ultrasonication must be weighed against the savings from milder reagents. They suggest that combining ultrasound with conventional agitation, rather than relying on it exclusively, may be the pragmatic path, and they call for pilot-scale trials with techno-economic analysis and life-cycle assessment. Still, the core message is striking: end-of-life solar panels are not merely a disposal problem but a concentrated, pre-sorted ore body, and with the right marriage of mechanical, thermal or solvent pre-treatment and green complexing acids—activated by nothing more exotic than sound—a recycler can dial in exactly which critical element to extract. As millions of tonnes of panels head toward retirement, that tunability may prove as valuable as the metals themselves.

Subject of Research: Green sonochemical recovery of technology-critical elements from end-of-life photovoltaic modules using aqueous organic-acid leaching solutions

Article Title: Green sonochemical recovery of technology-critical elements from the end-of-life photovoltaic module using aqueous organic-acid solutions

Article References: Green sonochemical recovery of technology-critical elements from the end-of-life photovoltaic module using aqueous organic-acid solutions. (n.d.). https://doi.org/10.1007/s10098-026-03606-5

Image Credits: AI Generated

DOI: 10.1007/s10098-026-03606-5

Keywords: photovoltaic recycling, sonochemical leaching, technology-critical elements, organic acids, urban mining, silver recovery, indium, tellurium, pyrolysis pretreatment, green chemistry, e-waste, circular economy