The core problem, according to the review, is complexity. HTC is not a single reaction but a tangled network of coupled processes, including dehydration, decarboxylation, polymerization, and recondensation, all unfolding simultaneously in an aqueous reactor. These transformations reshape the elemental composition, surface functionality, and microstructure of the starting biomass, which means that small changes in operating conditions can produce dramatically different materials. Feedstock heterogeneity compounds the challenge: two batches of the same agricultural waste can differ in moisture, ash content, and lignocellulosic composition, throwing off reproducibility between laboratories and even between runs in the same facility.

Design of experiments, a statistical discipline with roots in agricultural field trials and industrial quality control, offers a structured way to tame this complexity. Rather than varying one factor at a time, DoE frameworks such as factorial designs and response surface methodologies explore many variables simultaneously with statistically efficient experimental layouts. In the HTC context, the key variables include reaction temperature, residence time, the solid-to-liquid ratio in the reactor, system pressure, and the use of catalysts. By quantifying both the main effects of each variable and the interactions between them, DoE generates structured datasets that reveal which parameters matter most and how they combine to shape the final product.

Where DoE alone reaches its limits, machine learning picks up the baton. Classical response surface models fit relatively simple polynomial equations to experimental data, but HTC’s reaction networks are strongly nonlinear, with high-order interactions that polynomials struggle to capture. The review highlights how algorithms including artificial neural networks, ensemble methods, gradient boosting, and Bayesian approaches can learn these nonlinearities directly from data, substantially improving predictions of material-critical outputs. Among the most important targets are mass yield, the higher heating value, fixed carbon content, carbon retention, and textural and chemical descriptors that determine how well a hydrochar performs as an adsorbent or as a precursor for electrochemical materials.

The combination of the two approaches is more than the sum of its parts, the authors argue. DoE provides the statistically sound experimental backbone that ensures the training data for machine learning models actually covers the operating space of interest, while machine learning supplies the predictive muscle to capture relationships that classical models miss. This coupling proves particularly valuable for multi-objective optimization, one of the central tensions in hydrochar engineering. Energy densification, which maximizes heating value and fixed carbon, often works against the retention of oxygen-containing functional groups and the development of porosity, yet those very features determine whether a hydrochar can adsorb pollutants from water or serve as a scaffold for advanced carbon materials.

The choice of optimization target depends entirely on the intended application. Hydrochars destined for use as solid fuels benefit from conditions that push carbon content and calorific value upward, while chars intended for water and gas treatment require carefully preserved surface functionality and pore structure. A third family of applications treats hydrochar as a precursor for advanced carbon materials, where the priorities shift again. Because these objectives frequently conflict, the review emphasizes that DoE-guided machine learning allows researchers to map trade-off surfaces explicitly, identifying operating windows that balance competing demands rather than optimizing a single property in isolation.

Despite the promise, the review is candid about the bottlenecks that currently limit transferability and industrial robustness. Data quality and comparability across studies rank among the most serious. Machine learning models are only as good as the datasets they learn from, and HTC literature contains results generated under widely varying definitions, measurement protocols, and reporting standards. Pooling data across publications therefore risks training models on inconsistent inputs, a problem familiar to anyone who has tried to build predictive models from heterogeneous literature data in other fields of chemistry and materials science.

Interpretability presents a second hurdle. Deep neural networks and complex ensemble models may deliver impressive predictive accuracy, but they often behave as black boxes, offering little physical insight into why a particular combination of temperature, time, and feedstock produces a given outcome. For a field that still seeks mechanistic understanding of its reaction networks, models that cannot be interrogated risk becoming predictive oracles without explanatory power. The review suggests that hybrid strategies, in which DoE structure constrains and interprets machine learning predictions, offer a route to models that are both accurate and scientifically meaningful.

The systematic treatment of biomass variability rounds out the list of open challenges. Because feedstock composition varies with season, geography, and preprocessing, models trained on one biomass type may fail badly when applied to another. Building variability explicitly into experimental designs and model inputs, rather than treating it as noise, emerges as a methodological priority for anyone hoping to move hydrochar production from laboratory curiosity to industrial process.

Taken together, the review sketches a roadmap for moving HTC from empirical tuning toward data-informed, property-targeted engineering. By pairing statistically rigorous experimental design with machine learning capable of capturing high-order nonlinearities, researchers can begin to specify the hydrochar they need and work backward to the operating conditions that produce it. The authors outline methodological directions for developing more reliable hybrid DoE-ML strategies, with the goal of accelerating rational design of hydrochar materials at scale. If those strategies mature, the humble wet-biomass reactor could become a predictable manufacturing platform for carbon materials tailored to energy, environmental, and electrochemical applications alike.

Subject of Research: Design of experiments-guided machine learning for predicting hydrochar properties in hydrothermal carbonization

Article Title: Toward predictable hydrochar properties at scale: a critical review of design of experiments-guided machine learning in hydrothermal carbonization

Article References: El Ouadrhiri, F., & Lahkimi, A. (2026). Toward predictable hydrochar properties at scale: a critical review of design of experiments-guided machine learning in hydrothermal carbonization. Frontiers in Chemical Engineering, 8, Article 1883652. https://doi.org/10.3389/fceng.2026.1883652

Image Credits: AI Generated

DOI: 10.3389/fceng.2026.1883652

Keywords: hydrothermal carbonization, hydrochar, machine learning, design of experiments, biomass conversion, response surface methodology, neural networks, multi-objective optimization, higher heating value, adsorption, carbon materials, data-driven modeling

The headline result is striking. The material achieved a maximum fluoride adsorption capacity of 469.64 milligrams per gram, outperforming the magnesium oxide based adsorbents compared in the study. To put that figure in perspective, adsorption capacities in the tens of milligrams per gram are often considered respectable for fluoride removal, so a material approaching the 470 milligram per gram mark represents a substantial leap in performance. The researchers attribute this capacity to the intimate coupling of two components: a porous, plant-derived carbon scaffold and nanostructured magnesium oxide particles dispersed across its surface.

The synthesis route is deliberately simple, which matters enormously for any technology hoping to move from the laboratory bench to real treatment systems. The team first converted Rhus typhina biomass into hydrochar through hydrothermal carbonization, a process in which wet biomass is heated in water under pressure, producing a carbon-rich solid. This hydrochar was then subjected to pyrolysis in the presence of magnesium, yielding the final composite. The two-step pyro-hydro approach leaves behind a porous carbon architecture whose internal channels and cavities serve a dual purpose: they physically separate and stabilize the nano-MgO particles, preventing them from clumping together and losing reactive surface area, and they provide an enormous internal surface where fluoride ions can make contact with active sites.

Corresponding author Jiunian Guan framed the work as a deliberate double win. According to the research team, the goal was to address two environmental challenges simultaneously: fluoride contamination in water and the growing accumulation of invasive plant biomass. By converting the biomass into a functional carbon material and combining it with nano-magnesium oxide, the researchers created an adsorbent with strong fluoride-removal performance and promising environmental adaptability. That framing reflects a growing trend in environmental materials science, in which waste streams of one problem become feedstocks for solving another, closing loops rather than simply displacing pollution from one place to another.

The case for better fluoride treatment is urgent. Fluoride occurs naturally in groundwater, particularly in regions where fluoride-bearing minerals weather into aquifers, but human industry adds substantially to the load. Semiconductor manufacturing, glass production, electroplating, lithium refining, pesticide production, and pharmaceutical manufacturing all generate fluoride-containing wastewater. While trace fluoride strengthens tooth enamel and is deliberately added to municipal water in some countries, chronic exposure to elevated concentrations can cause dental fluorosis and, at higher doses, skeletal fluorosis, a painful and debilitating condition. Effective, affordable defluoridation technologies are therefore critical both for protecting drinking water supplies and for helping industry meet discharge standards.

One of the most practically important findings of the study is the material’s robustness across chemically diverse waters. Real wastewater is never clean; it carries a cocktail of dissolved salts whose ions compete with fluoride for adsorption sites. The nMgO/Py-HyC composite maintained stable fluoride adsorption across a broad pH range from 5.0 to 11.0, spanning acidic to strongly alkaline conditions. That breadth is unusual and valuable, because many adsorbents perform well only in a narrow pH window, forcing operators to add chemicals to precondition the water. Common coexisting ions such as nitrate and sulfate caused little interference with fluoride uptake, although the researchers did note that high concentrations of bicarbonate reduced adsorption, a limitation that will need to be considered when treating alkaline, bicarbonate-rich groundwaters.

Detailed material analyses revealed that fluoride capture is not the work of a single mechanism but a coordinated ensemble of processes operating in parallel. The identified mechanisms include electrostatic attraction, in which positively charged surface sites draw in the negatively charged fluoride ions; surface complexation, in which fluoride binds directly to magnesium centers at the surface; anion and ligand exchange, in which fluoride swaps places with hydroxyl groups or other ligands bound to the magnesium oxide; precipitation, in which dissolved fluoride reacts to form solid phases; and hydrogen bonding, which provides additional weak but numerous interactions. Spectroscopic and compositional evidence showed that fluoride could interact directly with magnesium to form stable compounds, including magnesium fluoride, known as MgF2, and fluorine-containing magnesium hydroxide phases. The involvement of precipitation and the formation of thermodynamically stable products is particularly significant, because it suggests that captured fluoride is locked in place rather than merely held loosely and at risk of leaching back out.

Perhaps the most scientifically elegant observation is the synergistic effect between the carbon support and the nano-magnesium oxide. Under identical experimental conditions, the adsorption capacity of the combined material was substantially greater than the sum of what the two components achieved separately. The porous carbon framework does more than passively hold the magnesium oxide; it exposes active magnesium sites that would otherwise be buried, improves the contact between fluoride ions in solution and those reactive sites, and may facilitate mass transport through its interconnected pore network. In materials chemistry, this kind of synergy, where the whole outperforms the parts, is the hallmark of a well-designed composite, and it is precisely what separates an engineered adsorbent from a simple mixture of ingredients.

Beyond the laboratory numbers, the study sketches out a circular strategy with genuine ecological appeal. Rhus typhina, commonly known as staghorn sumac, is an aggressive invader in many temperate regions, forming dense stands that displace native vegetation and require costly mechanical control. Harvesting its biomass and converting it into a high-value water treatment material transforms an ecological management burden into a resource, offsetting remediation costs while simultaneously cleaning contaminated water. If the supply chain can be organized, invasive plant management programs and water utilities could become unlikely partners, with the byproducts of habitat restoration feeding directly into drinking water protection.

The authors are appropriately measured about what remains to be done. Further work is needed to evaluate the material under different real-world wastewater conditions and at engineering scale, where factors such as column hydraulics, regeneration cycles, spent-adsorbent disposal, and long-term cost all come into play. A laboratory batch capacity of nearly 470 milligrams per gram must survive contact with turbid, variable, and chemically complex waters before it can be called a technology. Still, the findings provide a promising foundation for developing high-efficiency, low-carbon fluoride treatment technologies based on renewable biomass resources. In a field where adsorbents are often criticized as laboratory curiosities that never scale, a material that is cheap to make, works across a wide pH range, resists interference from common ions, and is built from an invasive species that needs removing anyway has an unusually credible path forward. The study, published in Biochar X with the DOI 10.48130/bchax-0026-0021, offers a template for how environmental chemistry, materials engineering, and invasive species management can converge on problems that no single discipline can solve alone.

Subject of Research: Nano-magnesium oxide modified pyro-hydrochar derived from invasive plant biomass for fluoride adsorption from water

Article Title: Turning invasive plants into powerful filters for fluoride-contaminated water

Article References: Turning invasive plants into powerful filters for fluoride-contaminated water. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: fluoride removal, adsorption, biochar, pyro-hydrochar, nano-magnesium oxide, Rhus typhina, invasive plants, water treatment, hydrochar, groundwater contamination, biomass utilization, environmental remediation

The study, published in the journal Results in Chemistry, was led by researchers at the Instituto Politécnico Nacional, working with material collected from a juice processing facility in Álamo, Veracruz, a state that produces nearly half of Mexico’s oranges. Their technique, hydrothermal carbonization, or HTC, subjects wet biomass to temperatures between 180 and 350 degrees Celsius under the pressure generated naturally by the sealed reactor. Unlike pyrolysis or gasification, HTC requires almost no energy to dry the feedstock, which makes it especially well suited to soggy agricultural residues like citrus peel. Water plays a double role during the process, acting simultaneously as solvent and catalyst, breaking apart polysaccharides, depolymerizing lignin, and stripping carbon dioxide from the molecular framework of the biomass.

The team dried and ground Valencia orange peels, pre-extracted them with ethanol, and then loaded ten grams of the prepared biomass into a one-liter stainless steel reactor with 100 milliliters of distilled water. After purging the vessel with nitrogen to create an inert atmosphere, they heated the mixture at a controlled rate of five degrees Celsius per minute to target temperatures ranging from 200 to 300 degrees Celsius, holding each run for 120 minutes under constant stirring. Autogenous pressures climbed from roughly 300 pounds per square inch at the low end to 1,300 at the highest temperature. The experiments were performed in duplicate, and every run yielded three products: a solid carbon-rich hydrochar, a gas composed mainly of carbon dioxide, and the prize the researchers were after, a liquid fraction they call light bio-oil.

Temperature proved decisive. Bio-oil yields climbed steadily from 8.9 percent of the dry feedstock mass at 200 degrees to a peak of 18.7 percent at 250 degrees, then fell back to 10.2 percent at 300 degrees. The pattern reflects a delicate chemical balancing act. At lower temperatures, hydrolysis of the lignocellulosic matrix remains limited, leaving cellulose and lignin largely intact. At 250 degrees, several productive reactions converge: polysaccharides break down into sugars that dehydrate into furan compounds, ether bonds within lignin cleave to release phenolic monomers such as guaiacol and vanillin derivatives, and cell walls rupture, liberating the flavonoids and phenolic acids native to citrus peel. Push the temperature higher, however, and secondary reactions take over. Reactive intermediates condense and repolymerize into heavy molecules that lock themselves into the hydrochar, while heat-sensitive flavonoids begin to degrade.

The resulting bio-oil, designated LBO-250, was a dark brown, viscous liquid with a density of 1.08 grams per milliliter, a pH of 4.2, and an unmistakable citrus aroma hinting at terpenes carried over from the peel’s essential oils. The acidity is typical of HTC products and stems from carboxylic acids formed during carbohydrate decomposition. To understand what the liquid actually contained, the researchers deployed a formidable analytical arsenal: scanning electron microscopy to track how the peel’s fibrous architecture collapsed during processing, ultraviolet-visible spectroscopy to detect the conjugated systems characteristic of flavonoids, thermogravimetric analysis to map the thermal stability of the solid residues, and infrared spectroscopy to identify the functional groups suspended in the oil.

The spectroscopic evidence converged on a consistent picture. Infrared spectra of LBO-250 showed a broad hydroxyl band near 3,400 wavenumbers, signatures of phenolic compounds, alongside carbonyl peaks at 1,736 wavenumbers pointing to carboxylic acids and esters, and a strong band at 1,636 wavenumbers indicating aromatic carbon-carbon double bonds. Nuclear magnetic resonance confirmed oxygenated carbons between 62 and 68 parts per million, aromatic and olefinic carbons between 120 and 155, and carboxyl carbons near 178, all hallmarks of lignocellulosic degradation products. Gas chromatography coupled with mass spectrometry then named the individual players: phenol, guaiacol, syringol, 4-ethylguaiacol, vanillin, guaiacylacetone, and cis-2-methoxycinnamic acid, a roster of well-documented chain-breaking antioxidants.

Quantitatively, the bio-oil contained 13.18 milligrams of gallic acid equivalents per gram in total phenolics, 71.54 milligrams of chlorogenic acid equivalents per gram of hydroxycinnamic acids, and 5.71 milligrams of quercetin equivalents per gram of flavonoids. Those numbers fall well short of Oligopin, a standardized commercial extract from French maritime pine bark, which delivered 443.25 milligrams of gallic acid equivalents per gram under the same assays. The gap is unsurprising. HTC operates well above the temperatures at which many plant polyphenols degrade, and orange peel contains only about 2 percent lignin, the principal renewable precursor of the aromatic phenols that HTC releases. The team’s earlier work on avocado peel, which is considerably richer in lignin, produced a bio-oil with far higher phenolic density under identical conditions.

The antioxidant testing revealed something more interesting than raw potency: radical selectivity. Against peroxyl radicals, the main chain-propagating species in lipid oxidation, LBO-250 achieved 25 percent inhibition at a concentration of 1,757.86 micrograms per milliliter, statistically significant but roughly five times weaker than Oligopin. Against hydroxyl and nitric oxide radicals, however, the bio-oil showed no measurable activity at all. The researchers attribute this not to a general weakness but to molecular architecture. Effective hydroxyl radical scavenging typically requires catechol or galloyl motifs capable of chelating transition metals, structures that were scarce among the methoxy-substituted phenols dominating the oil. Nitric oxide trapping, meanwhile, favors amino-bearing compounds or specific glycosylated flavonoids that were likewise absent. Peroxyl scavenging, by contrast, depends simply on hydrogen donation from phenolic hydroxyl groups, which the oil’s constituents supply readily.

That selectivity, the authors argue, defines rather than diminishes the product’s application space. Inhibiting lipid peroxidation is the dominant antioxidant function in food preservation, active packaging, and cosmetic formulations designed to protect against ultraviolet-induced oxidation, and a reproducible, radical-selective agent fits those niches well. Meanwhile, the co-produced hydrochar is far from waste: orange-peel-derived hydrochar has already been demonstrated as a precursor for activated carbons that capture carbon dioxide, and hydrochars generally serve as adsorbents, catalyst supports, and soil amendments. Continuous pilot-scale HTC of orange peel has shown the process can be scaled. The Mexican team, funded by national research programs, frames the work within the United Nations Sustainable Development Goals on energy, responsible production, and climate action, and calls for future optimization of temperature, pressure, and residence time to broaden the oil’s radical coverage. What began as a landfill problem may yet end up on the ingredient label of a cosmetic cream, or sealed inside a carbon filter, a small but tangible example of chemistry turning agricultural refuse into resource.

Subject of Research: Hydrothermal carbonization of orange peel waste to produce antioxidant bio-oil fractions

Article Title: Hydrothermal carbonization of orange peel waste: physicochemical characterization and antioxidant activity evaluation of bio-oil fractions

Article References: Rodríguez-Castillo, L. J., Méndez, F. J., García-Macedo, J. A., Calzada, L. A., Albíter, E., García-Pérez, M. E., Avalos-Viveros, M., de los Reyes, J., Santolalla-Vargas, C. E., Gómez, E., & Santes, V. (2026). Hydrothermal carbonization of orange peel waste: physicochemical characterization and antioxidant activity evaluation of bio-oil fractions. Results in Chemistry, 31, Article 103928. https://doi.org/10.1016/j.rechem.2026.103928

Image Credits: AI Generated

DOI: 10.1016/j.rechem.2026.103928

Keywords: hydrothermal carbonization, orange peel waste, bio-oil, antioxidants, phenolic compounds, flavonoids, hydrochar, biomass valorization, circular economy, green chemistry, Citrus sinensis, radical scavenging

The review, led by Pengfei Xu and colleagues, analyzed more than 61,000 publications indexed in the Web of Science Core Collection between 2000 and 2025, of which 7,893 focused specifically on soil carbon sequestration. The bibliometric analysis reveals a research field that has grown explosively, with publication output accelerating sharply after the 2016 Paris Agreement. But it also exposes a striking imbalance: outcome-related keywords such as soil organic carbon and greenhouse gas emissions dominate the literature, while process-oriented terms like microbial community, priming effect and adsorption remain far less visible. The authors argue that this gap between what scientists measure and what they explain has left the field without a unified mechanistic framework.

Artificial carbon materials come in three main flavors, each forged from biomass through different thermochemical routes. Biochar, the best studied of the three, is produced by heating biomass above 300 degrees Celsius under oxygen-limited conditions, yielding a carbon-rich, aromatic material. Hydrochar is made through hydrothermal carbonization, which uses water and moderate temperatures and produces a more oxygenated, more microbially accessible substance. Artificial humic substances, the youngest class, are created through alkaline hydrothermal humification, a process that accelerates the chemistry of natural humification and yields oxygen-rich materials resembling natural soil humic acids. The review is careful to keep these classes distinct, noting that they differ profoundly in molecular composition, surface chemistry and likely behavior in soil.

The evidence that these materials raise measured soil organic carbon is generally positive but strikingly heterogeneous. In one 180-day incubation of maize cropland topsoil in Jilin, China, cotton straw-derived biochar applied at equivalent carbon inputs increased soil organic carbon concentration by 94.6 percent, compared with 64.4 percent for raw straw. In a paddy soil column experiment, hydrochar applications increased native soil organic carbon by 1.0 to 3.0 percent while shifting carbon from labile to stable pools. In black soil from Harbin, artificial humic acid raised total organic carbon by 21.4 grams per kilogram within 28 days. But the authors caution that these numbers cannot be pooled into a universal average, because responses depend on feedstock, production conditions, soil texture, mineralogy, pH, moisture, climate and experimental duration.

The most provocative part of the review concerns what a rise in soil carbon actually means for the climate. Soils globally release roughly 60 petagrams of carbon per year through decomposition, a gross flux that dwarfs annual fossil fuel emissions, though it is largely part of the natural carbon cycle. When an amendment is added, total carbon dioxide emissions may fall or rise, but a bulk flux measurement alone cannot tell scientists whether the carbon came from the amendment or from the soil’s native organic matter. The review draws a sharp evidentiary line: only isotope-based or otherwise source-resolved methods count as direct evidence of priming, the acceleration or suppression of native soil carbon decomposition. Many widely cited studies, it turns out, fall short of that standard.

Biochar studies report both positive and negative priming depending on pyrolysis conditions, soil pH and texture, substrate supply and duration. Hydrochar, with its soluble and aliphatic components, may boost short-term respiration, but the authors stress this is not proof of positive priming without source separation. Evidence for artificial humic substances remains thin, with few isotope-tracing studies. The message is blunt: no material class can be assumed to uniformly suppress native carbon loss, and short-term declines in carbon dioxide emissions may simply reflect transient microbial suppression or substrate limitation rather than genuine stabilization.

The mechanistic heart of the review organizes sequestration into three interacting pillars. The first is material persistence: pyrolysis creates condensed aromatic structures that resist decomposition, while hydrochar contains larger microbially accessible fractions, and artificial humic substances may gain persistence through association with minerals despite their chemical reactivity. Elemental ratios such as hydrogen-to-carbon and oxygen-to-carbon serve as rough screening tools, but the authors warn they cannot capture pore accessibility, mineral coatings or carbon-domain heterogeneity. Long-term isotope studies confirm that biochar contains persistent fractions while also releasing a measurable labile fraction shortly after incorporation, yet comparable multi-year field data for the other two classes are almost entirely absent.

The second pillar is physical and mineral protection. Soil aggregates restrict microbial access to carbon, and artificial carbon materials can promote aggregation through their pores, surface chemistry and interactions with clay minerals, roots, fungal hyphae and microbial glues. In grassland and rice paddy soils, biochar increased macroaggregates within the first 60 days, with microaggregates enriched in biochar-derived carbon developing later. Mineral type matters enormously: expanding 2:1 clays like montmorillonite offer far more sorption capacity than kaolinite, and iron and aluminum oxides can lock up carboxyl-rich compounds through ligand exchange and co-precipitation. Aging further reshapes these interfaces, with mineral bonding detected on biochar after roughly three months and organo-mineral coatings observed on particles recovered from field trials after up to 9.5 years.

The third pillar is microbial transformation. Artificial carbon materials modify habitat conditions, buffering water fluctuations and altering pH, nutrients and redox, while their soluble compounds feed microbes. Rather than simply sitting inert, amendment carbon can flow through microbial biomass into necromass, the dead residues that subsequently bind to minerals or become sealed inside aggregates. A 12-year field experiment found biochar increased microbial necromass carbon by 23.3 to 39.0 percent in topsoil but reduced it at depth, while a four-year paddy study reported decreases at all depths measured, a reminder that total soil carbon gains can coexist with losses in specific microbial carbon pools. Microspectroscopic work in a long-term pasture showed biochar particles sorbing root-derived carbon and accumulating mineral coatings that incorporated microbial necromass, effectively lifting the soil’s carbon ceiling.

The review’s practical upshot is that there is no universally optimal material. Alkaline biochars suit acidic soils but risk over-liming neutral ones; hydrochar fits wet wastes that need no pre-drying but carries soluble carbon whose fate varies; artificial humic substances offer reactive mineral interactions and may even promote biological carbon dioxide assimilation, though their long-term field persistence is unproven. The authors call for multi-year, multi-site trials using isotope source partitioning, standardized aging protocols, integrated mineralogical and microbial analyses, and full life-cycle assessments. As the 4 per mille initiative seeks to grow global soil carbon stocks by 0.4 percent annually, this review offers both a caution and a roadmap: engineered carbon materials can genuinely help soils fight climate change, but only if scientists and farmers track where the carbon comes from, where it goes, and how long it stays.

Subject of Research: Mechanisms of soil organic carbon sequestration by artificial carbon materials in agroecosystems

Article Title: Artificial carbon materials for soil organic carbon sequestration in agroecosystems: A bibliometric analysis and mechanistic review

Article References: Xu, P., Yu, H., Zhang, J., Xu, Y., & Liu, Z. (2026). Artificial carbon materials for soil organic carbon sequestration in agroecosystems: A bibliometric analysis and mechanistic review. Journal of Agriculture and Food Research, 31, Article 103343. https://doi.org/10.1016/j.jafr.2026.103343

Image Credits: AI Generated

DOI: 10.1016/j.jafr.2026.103343

Keywords: soil organic carbon, biochar, hydrochar, artificial humic substances, carbon sequestration, priming effect, soil aggregates, mineral association, microbial necromass, greenhouse gas emissions, agroecosystems, climate change mitigation

The team’s starting point was a gap in the literature. The link between soil aggregates and soil organic carbon cycling has been documented extensively, yet the combined effects of irrigation regime and exogenous organic carbon additions on aggregate distribution, aggregate-bound carbon and microbial communities in paddy soils remained poorly understood. Paddy soils are an unusual case: they spend much of the growing season saturated, which changes everything from oxygen availability to how particles stick together. As water-saving irrigation schemes spread across rice-growing regions to conserve scarce freshwater, researchers have worried about what drier, more fluctuating moisture conditions do to the physical structures that protect carbon in these soils. The new experiment was designed to answer that question directly.

Methodologically, the study was deliberately gentle with its samples. Rather than subjecting soil to aggressive dry sieving, which can shatter natural aggregates and distort the size distribution, the researchers used an optimal-moisture sieving method, working the soil at a moisture content chosen to preserve aggregate integrity. Soil was separated into size classes, from large aggregates larger than 2 millimeters down to fine fractions smaller than 0.25 millimeters. The treatments compared flooding irrigation with controlled irrigation, and each water regime was crossed with additions of rice straw or with hydrochar produced from that same straw by hydrothermal carbonization — a process that converts wet biomass into a carbon-rich, chemically altered solid using heat and pressure in water.

The first striking result is how lopsided paddy soil architecture turned out to be. Across the treatments, aggregates larger than 2 millimeters dominated the soil mass, accounting for between 67 and 80 percent of the total, while the finest fraction below 0.25 millimeters was the least abundant class at just 7 to 16 percent. That imbalance matters more than it might seem. When scientists calculate how much organic carbon sits in the soil as a whole, the bulk of the number comes from whatever size class holds most of the mass — in this case, the big aggregates — even if smaller fractions are individually richer in carbon per gram. The study confirmed exactly this pattern: the fine fractions were enriched in organic carbon and microbial attributes, yet the larger-than-2-millimeter class made the largest numerical contribution to calculated bulk soil organic carbon simply because of its overwhelming mass proportion.

Water management left a clear fingerprint on this architecture. Compared with continuous flooding, controlled irrigation increased the proportion of large aggregates greater than 2 millimeters and improved aggregate-related structural indices, as reflected in higher values of mean weight diameter and geometric mean diameter — two standard measures that essentially summarize whether a soil is dominated by sturdy, well-formed crumbs or by loose, easily eroded fragments. For a water-saving technique, that is a reassuring outcome. It suggests that the drying and re-wetting cycles inherent to controlled irrigation do not necessarily degrade paddy soil structure; instead, under the conditions of this experiment, they appear to have nudged the soil toward a coarser, better-aggregated state.

Adding exogenous organic carbon raised the amount of organic carbon associated with aggregates in both water regimes, but the two amendments behaved in intriguingly different ways. Rice straw proved the stronger stimulator of microbial biomass, consistent with its role as a fresh, easily decomposable food source for soil organisms. Hydrochar, by contrast, produced a greater increase in the proportion of large aggregates and in the organic carbon concentration within that greater-than-2-millimeter class. In other words, straw fed the microbes, while hydrochar built the structure — and locked more carbon into the biggest, mass-dominant crumbs. The authors conclude that hydrochar amendment under controlled irrigation improved aggregate-related structural characteristics and increased aggregate-associated organic carbon mainly by boosting both the mass share and the carbon concentration of the large-aggregate fraction.

The interaction between the two levers was not simply additive. The study found that controlled irrigation and hydrochar amendment influenced aggregate distribution and aggregate-associated organic carbon independently and, for some aggregate-size responses, interactively. This nuance carries practical weight: it means farmers and researchers cannot assume that the effect of a soil amendment is the same under a flooded field as under a water-saving schedule. The moisture regime sets a context that shapes how organic additions translate into physical structure and carbon stabilization, which is precisely the kind of information needed to design management packages rather than isolated interventions.

Why should carbon cling to large aggregates so effectively when hydrochar enters the picture? The mechanistic story, as the authors and the surrounding literature frame it, involves the way particulate organic materials act as binding agents. Fresh plant residues serve as nuclei around which mineral particles and microbial products accumulate, forming larger and more stable aggregates. Hydrochar, being more chemically recalcitrant than raw straw, persists longer and may act as a durable skeletal component within these crumbs, while its own carbon rides along inside the aggregate structure. Meanwhile, the enrichment of organic carbon and microbial attributes in the fine fractions reflects the reality that microaggregates offer protected microhabitats where decomposers and their substrates are held in close, moisture-buffered contact. Both ends of the size spectrum do important work — they just do it in different currencies.

For rice systems, which cover vast areas of Asia and store substantial carbon in their periodically waterlogged soils, the findings arrive at a moment when water scarcity is pushing irrigation reform and carbon accounting is tightening around agriculture. Controlled irrigation is already valued for cutting water use and, according to related work by overlapping research groups, for mitigating methane emissions from paddies. This study adds a structural dimension to its credentials: the practice did not sacrifice, and in fact improved, the aggregation metrics that underpin soil physical health. Pairing it with hydrochar made from the very straw that rice harvests generate offers a circular route — crop residue is converted off-field and returned as a stable carbon amendment that reinforces soil structure rather than decomposing rapidly and releasing its carbon back to the air.

Cautions remain, as they always do with short-term experiments. The aggregates and microbial responses reported here were measured under experimental conditions, and longer-term field studies will be needed to confirm that hydrochar-driven gains in large-aggregate carbon persist across seasons and soil types, and that microbial stimulation by straw does not simply accelerate carbon loss elsewhere in the system. The researchers also note that datasets from the study are available from the corresponding author on reasonable request, inviting replication. Still, the central message is crisp and actionable: the amount of carbon a paddy soil can bank depends not just on what you add to it, but on how wet you keep it — and the best results come when the two are tuned together.

Subject of Research: Effects of controlled irrigation and hydrochar amendment on soil aggregate-size distribution and aggregate-associated organic carbon in paddy soils

Article Title: Controlled irrigation and hydrocchar amendment alter aggregate-size distribution and aggregate-associated organic carbon in paddy soils

Article References: Wang, K., Xu, J., Chen, P., Fan, J., Hu, J., Chang, Y., & Sleutel, S. (2026). Controlled irrigation and hydrocchar amendment alter aggregate-size distribution and aggregate-associated organic carbon in paddy soils. Plant and Soil. https://doi.org/10.1007/s11104-026-09144-9

Image Credits: AI Generated

DOI: 10.1007/s11104-026-09144-9

Keywords: paddy soil, soil aggregates, hydrochar, controlled irrigation, soil organic carbon, rice straw, microbial biomass, aggregate stability, carbon sequestration, water-saving irrigation, soil structure, hydrothermal carbonization

The research team, led by Zuzanna Prus-Frączek and including senior authors Katarzyna Styszko and Małgorzata Wilk, focused on polyethylene terephthalate, or PET, the polymer of water bottles and food packaging, and polystyrene, or PS, the aromatic plastic found in insulation, packaging foam, and countless disposable items. Both are among the most common polymers recovered from treated sludge, which can harbor hundreds to thousands of microplastic particles per kilogram of dry matter. Because European Union policy increasingly promotes the reuse of sludge as fertilizer and soil amendment on farmland, understanding whether hydrothermal treatment truly breaks these particles down is not an academic detail. It is central to whether plastic pollution quietly rides the circular economy back into the food chain.

Earlier studies had reported that hydrothermal carbonization can slash the number of microplastic particles detectable in sludge by up to 90 percent, with one study showing removal of particles larger than 50 micrometers jumping from about 11 percent to 85 percent as temperatures rose from 170 to 220 degrees Celsius. But a reduction in detectable particles is not the same as destruction. Fragments can shatter below the size limits of analytical instruments, and polymer-derived compounds can dissolve into the surrounding water without ever being mineralized. The Polish team set out to determine which of these fates actually befalls PET and PS, treating the two polymers separately in pure water at 220 degrees Celsius for one, two, and three hours, conditions carefully chosen to mirror real sewage sludge processing while eliminating interference from sludge-derived organic matter.

The results revealed a striking asymmetry between the two plastics. PET, whose backbone contains ester bonds that water can chemically attack, proved far more vulnerable. Thermogravimetric analysis showed that its degradation onset temperature, which stands at roughly 402 degrees Celsius in the raw polymer, plummeted to between 246 and 284 degrees Celsius after treatment. The pyrolytic residue left at 800 degrees Celsius collapsed from 12.7 percent to as little as 1.3 percent, indicating a severely diminished tendency to form stable carbon structures. Infrared and Raman spectroscopy painted a consistent picture: the sharp ester carbonyl band at 1714 inverse centimeters gave way to new dominant components near 1671 to 1677 inverse centimeters, signaling a heterogeneous mix of partially hydrolyzed ester bonds and newly formed oxygen-containing structures. Scanning electron microscopy showed surfaces transformed from smooth and compact into rough, cracked, and cavity-riddled landscapes.

Polystyrene, by contrast, emerged from the reactor remarkably intact. Its hydrophobic aromatic skeleton, held together by carbon-carbon bonds that water cannot easily cleave, resisted the conditions that visibly weakened PET. Treated PS samples retained a single-step thermal degradation profile with maximum decomposition temperatures of around 416 to 417 degrees Celsius, essentially unchanged from the raw material. The principal FTIR and Raman bands, including the characteristic aromatic ring breathing mode near 1001 inverse centimeters, remained within measurement uncertainty of their original positions. Only slight band broadening after three hours hinted at limited local disorder at the surface. The finding aligns with earlier work showing that complete PS decomposition demands 360 degrees Celsius and four hours, conditions far more severe than any sewage sludge reactor would employ.

Perhaps the most consequential discovery came from analyzing the process water itself. When the researchers measured chemical oxygen demand and total organic carbon in the liquid phase, PET proved to be a prodigious polluter. Normalized to the initial polymer mass, PET released approximately 737 to 876 milligrams of COD and 320 to 343 milligrams of TOC per gram of plastic into the water, with absolute concentrations reaching 6145 to 7300 milligrams per liter of COD and around 2700 to 2855 milligrams per liter of TOC. Polystyrene, in comparison, contributed only 35 to 83 milligrams of COD and 99 to 140 milligrams of TOC per gram of polymer. The pH of the PET process water remained stubbornly acidic, between 3.8 and 3.9, consistent with the accumulation of acidic products from ester bond cleavage, while phenolic compounds slowly increased with longer treatment, suggesting the formation of refractory aromatic derivatives that resist further degradation.

These numbers carry a direct message for engineers designing full-scale sludge hydrothermal carbonization plants. Process water from sludge HTC is already heavily loaded with dissolved organics and can contain inhibitory or toxic compounds, and adding a substantial polymer-derived burden, particularly from PET-rich sludge, would amplify the challenge of treating or safely recirculating that water. Previous pilot-scale studies have shown that biological treatment of HTC process water can achieve COD removals of roughly 71 to 75 percent in membrane bioreactors and sequencing batch reactors, but the researchers caution that the added plastic-derived load could strain these systems further. Hydrochar yields added another layer of nuance: the most intensive apparent conversion for both polymers occurred at two hours, with yields for PS dropping to 75.6 percent and PET to 53.6 percent, and the non-monotonic trends hint at competing processes of depolymerization, dissolution, and secondary condensation of soluble intermediates back onto the solid phase.

The environmental implications extend beyond the treatment plant. Hydrochar from PET-containing sludge, with its weakened structure and earlier degradation onset, would likely be more susceptible to further fragmentation, oxidation, and weathering if applied to soils, potentially generating new generations of smaller plastic particles. The dramatic reduction in pyrolytic residue also suggests a diminished capacity for stable carbon sequestration in PET-derived char, undermining one of the climate arguments for hydrochar use in agriculture. Polystyrene-derived material, meanwhile, may remain embedded within or incorporated into the carbonized matrix despite surface alteration, retaining its aromatic identity. In neither case does the plastic vanish. As the authors put it, HTC should be understood primarily as a polymer-transformation and redistribution process, shifting material between the solid, liquid, and gas phases rather than eliminating it from the system.

The study does have limits that the team acknowledges candidly. The experiments used pristine, manually fragmented waste plastics in distilled water at a single temperature, rather than environmentally aged particles in a real sludge matrix, and the analyses could not identify specific molecular transformation products, distinguish hydrolysis from oxidation, or assess nanoplastic formation because samples were ground before imaging. Future work, the researchers argue, should incorporate real sewage sludge, replicate experiments across broader operating windows, and deploy gas and liquid chromatography coupled with mass spectrometry to identify individual compounds, alongside leaching and ecotoxicity tests to evaluate the fate of released organics in soils. For now, the message for the water sector is clear and somewhat uncomfortable: hydrothermal carbonization remains a valuable technology for valorizing wet sludge, but claims of microplastic removal must be read as claims of microplastic transformation, and any credible plant design must treat the process water as a first-class waste stream, not an afterthought.

Subject of Research: Hydrothermal carbonization of PET and polystyrene microplastics under sewage sludge treatment conditions

Article Title: Hydrothermal carbonization of PET and PS microplastics under conditions relevant to sewage sludge treatment

Article References: Prus-Frączek, Z., Kalemba-Rec, I., Magdziarz, A., Dróżdż, A., Chwiej, J., Styszko, K., & Wilk, M. (2026). Hydrothermal carbonization of PET and PS microplastics under conditions relevant to sewage sludge treatment. Energy Reports, 16, Article 109697. https://doi.org/10.1016/j.egyr.2026.109697

Image Credits: AI Generated

DOI: 10.1016/j.egyr.2026.109697

Keywords: microplastics, hydrothermal carbonization, sewage sludge, PET, polystyrene, hydrochar, process water, wastewater treatment, COD, TOC, polymer degradation, circular economy