Biochar-based approaches for heavy metal remediation in agricultural soils: Mechanisms, optimization, and emerging AI applications

Abstract

Heavy metal contamination in agricultural soils poses persistent risks to crop safety and food-chain exposure. Although biochar is widely proposed—and increasingly applied—as a remediation amendment, field performance remains highly variable across soil constraints, metal speciation, and biochar designs. This review addresses this uncertainty by translating immobilization pathways (sorption/ complexation, precipitation, and redox-mediated stabilization) into a decision-oriented “mechanism–lever–endpoint” framework, thus linking mechanistic hypotheses to controllable engineering strategies such as feedstock selection, pyrolysis windows, and mineral/composite design. Beyond established plant–microbe interactions, there is a critical assessment of under-synthesized biochar–soil fauna pathways, with a focus on earthworms, and a reconciliation of conflicting evidence by highlighting boundary conditions that shift biological responses. Agronomic trade-offs and environmental risks are considered, associated with biochar production and application, emphasizing failure modes relevant to long-term soil health and remediation reliability. To support decision-grade deployment under heterogeneous evidence, a bias-aware AI-assisted workflow is outlined, which stresses standardized reporting, interpretability, and leakage-safe validation. Overall, the review integrates engineering options with biological synergies into a practical roadmap for more predictable and site-specific remediation in agricultural soils.

Full Article

Biochar-Based Approaches for Heavy Metal Remediation in Agricultural Soils: Mechanisms, Optimization, and Emerging AI Applications

Rui Wu,^a † Mingxian Zhang, ^c † Faidzul Hakim Adnan,^a Pei Yi Siow  ,^b and Mohd Izzudin Izzat Zainal Abidin  ^a,*

Heavy metal contamination in agricultural soils poses persistent risks to crop safety and food-chain exposure. Although biochar is widely proposed—and increasingly applied—as a remediation amendment, field performance remains highly variable across soil constraints, metal speciation, and biochar designs. This review addresses this uncertainty by translating immobilization pathways (sorption/ complexation, precipitation, and redox-mediated stabilization) into a decision-oriented “mechanism–lever–endpoint” framework, thus linking mechanistic hypotheses to controllable engineering strategies such as feedstock selection, pyrolysis windows, and mineral/composite design. Beyond established plant–microbe interactions, there is a critical assessment of under-synthesized biochar–soil fauna pathways, with a focus on earthworms, and a reconciliation of conflicting evidence by highlighting boundary conditions that shift biological responses. Agronomic trade-offs and environmental risks are considered, associated with biochar production and application, emphasizing failure modes relevant to long-term soil health and remediation reliability. To support decision-grade deployment under heterogeneous evidence, a bias-aware AI-assisted workflow is outlined, which stresses standardized reporting, interpretability, and leakage-safe validation. Overall, the review integrates engineering options with biological synergies into a practical roadmap for more predictable and site-specific remediation in agricultural soils.

DOI: 10.15376/biores.21.3.Wu

Keywords: Biochar; Agricultural soils; Heavy-metal immobilization; Modification strategies; Machine learning (ML)

Contact information: a: Sustainable Process Engineering Centre (SPEC), Department of Chemical Engineering, Faculty of Engineering, Universiti Malaya, 50603 Kuala Lumpur, Malaysia; b: Department of Mechanical Engineering, Faculty of Engineering, Universiti Malaya, 50603, Kuala Lumpur, Malaysia; c: School of Civil Engineering and Urban Planning, Liupanshui Normal University, Liupanshui 553004, China; † These authors contributed equally to this work;* Corresponding author: Mohd Izzudin Izzat Zainal Abidin; izzudinizzat@um.edu.my

Graphical Abstract

INTRODUCTION

Heavy-metal contamination in agricultural soils remains a persistent food-safety challenge because risk is governed not only by total metal concentration, but more directly by metal bioavailability at the soil–plant interface and subsequent transfer along the food chain (Lawal et al. 2021). Unlike natural geochemical backgrounds, modern soil pollution is increasingly dominated by heterogeneous anthropogenic inputs—including mining, smelting, wastewater irrigation, and agrochemical use—thereby creating contamination profiles that rarely respond to uniform remediation strategies (Fei et al. 2022; Qin et al. 2021). Consequently, controlling bioavailability has become as important as reducing total metal concentrations. Remediation strategies that specifically minimize bioavailability and maintain long-term stability under variable soil conditions are especially needed.

Conventional organic amendments such as compost and manure are widely used in agricultural soils, but their remediation performance may be limited by rapid decomposition, variable composition, and, in some cases, dissolved organic carbon (DOC)-mediated metal mobilization (D. Chen et al. 2022; Xu et al. 2025). In this context, biochar has progressed from a general soil amendment to an increasingly engineerable material platform for farmland remediation (Aziz and Kareem 2023). Unlike more labile amendments, biochar is a thermochemically transformed carbonaceous material whose pore architecture, surface chemistry, alkalinity, and mineral associations can be more deliberately tuned through feedstock selection, pyrolysis conditions, and post-modification. Its remediation efficacy arises from coupled interactions—including surface complexation, precipitation, and, in some cases, redox-mediated stabilization—making it particularly relevant for heavy-metal immobilization under diverse soil constraints (Ji et al. 2022; Gusiatin and Rouhani, 2023). A central research gap persists. There needs to be a transition from trial-and-error experimentation to rational, context-specific biochar design. Despite a rapidly expanding literature, biochar engineering is still frequently pursued through ad hoc testing, and many existing reviews emphasize either synthesis routes or broad application benefits without clearly identifying which engineering lever is most appropriate under specific soil–metal scenarios (Venkatachalam et al. 2023). Much less attention has been paid to a practical decision framework that explicitly links mechanistic soil cues to controllable engineering levers. Without such a framework, the field risks remain reliant on fragmented case-by-case evidence with limited predictive power.

Artificial intelligence (AI) is increasingly being explored as a decision-support layer for decoding high-dimensional interactions among pyrolysis parameters, soil properties, and immobilization outcomes (Nguyen et al. 2024). However, the reliability of AI in this domain remains constrained by fragmented reporting, data bias, weak validation practices, and the inclusion of descriptors that are not always directly aligned with remediation goals. Accordingly, this review advances a mechanism-centered and decision-oriented perspective on biochar-based heavy-metal remediation. Specifically, the authors (1) develop a decision-oriented framework linking immobilization mechanisms to controllable engineering levers across soil and metal contexts; (2) evaluate integrated biochar–plant–microbe strategies by identifying synergy pathways and boundary conditions; (3) critically assess the role of AI in biochar design and remediation prediction, with emphasis on remediation-relevant variables and validation challenges; and (4) highlight risk-aware deployment, arguing that success should be judged not only by short-term immobilization, but also by long-term stability and ecological safety. This review is therefore structured from mechanistic soil controls to biochar design and optimization, and finally to AI-assisted decision support for more predictable remediation outcomes.

SOURCES, PATHWAYS, AND BIOAVAILABILITY OF HEAVY METALS IN AGRICULTURAL SOILS

Understanding the sources, transformation pathways, and bioavailability of heavy metals is essential for rational biochar-based remediation. This is because these controls determine which immobilization mechanisms and material properties are most appropriate under specific soil conditions. Heavy metals contamination in farmlands seldom stems from a single origin (Fig 1). Most sites exhibit mixed-source legacies in which geogenic backgrounds intersect with anthropogenic inputs (Table S1). In practice, contamination is assessed against local geochemical backgrounds or regulatory thresholds, not a universal rule (Rothwell and Cooke 2015). Attention typically centers on topsoil, which exchanges rapidly with air and water and directly interfaces with crop roots (O’Connor et al. 2018). This context makes it necessary to understand metal provenance, transport, and bioavailability before selecting biochar-based remediation strategies intended to deliver predictable field performance.

Fig. 1. Major sources and pathways of heavy metals into agricultural soils (Zhao et al. 2025, under CC BY 4.0 license)

Sources and Entry Routes

Natural inputs arise from weathering of parent materials, volcanic and aeolian deposition, and lithological anomalies (Zhang and Wang 2020). While natural processes establish background levels and can locally contribute substantially, anthropogenic sources generally dominate contemporary enrichment patterns. Major human-derived inputs include mining and smelting, wastewater irrigation, agrochemical applications (fertilizers, pesticides, fungicides with trace-metal impurities), organic amendments such as manure and sewage sludge, industrial emissions, traffic, and mismanaged solid wastes (Joshi and Gururani 2023). Once released, metals are transferred to agricultural soils through multiple pathways, atmospheric deposition (Qiao et al. 2023), irrigation with contaminated water (Joshi and Gururani 2023), runoff and leaching from tailings or waste sites, and direct inputs through fertilizers and amendments (Zhang and Wang 2020). The relative contribution of these pathways varies across regions and land-use histories, yet their convergence commonly results in cumulative and persistent enrichment in farmland soils. This source heterogeneity partly explains why uniform amendment strategies often perform inconsistently across sites and why remediation design must remain context-specific.

Transformations and Controlling Factors in Soils

After entering soils, metals repartition among soil solution, exchange sites, and more specific associations such as surface complexation on Fe/Al/Mn (hydr)oxides and co-precipitation with carbonates or phosphates (Dhiman et al. 2023). In agricultural systems, a biologically mediated fraction may also be transferred into plant tissues, from which metals can later re-enter the soil through residue return and decomposition (Tan et al. 2023). The pH remains the primary regulator of cationic metal mobility (Zhang et al. 2022), while redox potential (Eh) dictates speciation and the stability of Fe/Mn phases; under flooded or reducing conditions Fe/Mn oxide dissolution can release previously sorbed ions and convert As(V) to the more mobile As(III) (Yu et al. 2024). In addition to these inorganic controls, dissolved organic carbon (DOC) often complexes Cu and other trace metals, sometimes enhancing mobility. Over time, aging and occlusion processes progressively reduce lability as ions diffuse into micro-/mesopores or become encapsulated within recrystallizing minerals (Li and Gong 2021). Ultimately, total metal concentrations are often poor predictors of environmental risk; instead, risk depends more directly on the fraction that remains chemically labile and biologically accessible. These physicochemical controls also define the design requirements for remediation, including whether performance is more likely to depend on alkalinity, mineral-assisted sorption, redox buffering, or suppression of DOC-mediated mobilization.

Bioavailability: Concept and Measurement

Bioavailability denotes the fraction of a metal that is accessible for biological uptake over relevant timescales (Xu et al. 2022); it is method- and organism-dependent. Operational proxies interrogate different processes. Researchers can choose between simple extractions (e.g., CaCl₂, NH₄NO₃, DTPA/EDTA), sequential extraction schemes that partition operational pools, diffusive gradients in thin films (DGT) for labile supply, and direct porewater or isotopic-exchange approaches (O’Connor et al. 2018; Zhu et al. 2023). Method choice should be decision-oriented and locally calibrated to soils and crops; for example, organic amendments such as compost can lower plant burdens mainly by shrinking the labile pool, not by changing totals (D. Chen et al. 2022). Linking measurement to management is therefore essential, because remediation should target the chemical pool that actually drives ecological exposure and food-chain risk. For biochar-based remediation in particular, success should be judged mainly by reductions in labile and biologically relevant pools rather than by changes in total concentrations alone.

Implications for Risk and Management

Elevated labile fractions of heavy metals can impair seed germination, plant growth, photosynthesis, and cellular homeostasis in crops and pose human-health risks through food-chain transfer (Angon et al. 2024; Rai et al. 2023). Source control remains the primary mitigation strategy wherever feasible. However, when in-situ remediation is required, effective intervention depends on aligning material choice with the soil’s governing chemical constraints rather than relying on trial-and-error amendment application. In practical terms, the mobility-limiting factors of different contaminants dictate the required remediation functions: acidic soils contaminated with cationic metals generally require stronger pH buffering and precipitation-promoting capacity, redox-dynamic environments require greater stabilization against speciation shifts, and DOC-rich systems demand materials capable of limiting ligand-assisted remobilization. These site-specific cues provide the conceptual basis for the mechanism–lever–endpoint framework developed in the next section. By linking the soil constraints summarized here to the engineered biochar properties and immobilization pathways discussed later, remediation strategies can be designed in a more predictable and field-relevant manner.

ENGINEERING BIOCHAR FOR HEAVY-METAL REMEDIATION: PROPERTIES, MODIFICATION LEVERS, AND PROCESS DESIGN

Biochar performance in agricultural soils is governed by engineerable physicochemical attributes—pore architecture, surface functional chemistry, and alkalinity/surface charge—that map onto the dominant immobilization pathways in soils (sorption, surface complexation/co-precipitation, and redox interactions) (Ibrahim et al. 2022). Feedstock and thermochemical history set the baseline, whereas post-synthesis modification extends the design space for site-specific remediation. A process view is therefore essential: from soil cues, to property targets, modification choices, and field-ready materials and metrics.

Pore Architecture and Accessible Surface

Physical sorption contributes meaningfully when the pore network provides both capacity (internal surface) and accessibility (mass-transfer pathways) (Hou et al. 2024; Xing et al. 2023). The IUPAC classification—micropores (< 2 nm), mesopores (2 to 50 nm), and macropores (> 50 nm)—is useful operationally: micropores often supply the majority of internal surface (capacity), mesopores facilitate diffusion toward microporous domains, and macropores support transport and biotic colonization (Leng et al. 2021). Critically, overly deep “dead-end” pores may trap pollutants without proportionate risk reduction, whereas connected mesopores improve utilization of internal sites (Krebsbach et al. 2023). Brunauer–Emmett–Teller (BET) surface-area values are often misleading; Sinha et al. (2019) reported that BET surface area, together with pore-size distribution, is necessary to interpret performance beyond a single ‘surface area’ value. According to comprehensive reviews on biochar porosity engineering (Li et al. 2016), and consistent with the empirical patterns summarized in Table S2, BET surface area generally increases with pyrolysis temperature, although substantial variability remains due to feedstock characteristics and processing conditions. Reported values can range from less than 10 m²/g in some low-temperature chars (below 350 °C) to more than 400 m²/g in high-temperature or physically activated biochars.

Pyrolysis window and feedstock co-determine these traits, and studies consistently show that increasing microporosity and accessible surface area enhances immobilization of target metals in soils (Souza et al. 2023; Yakout et al. 2015). However, gains in microporosity may trade off with accessibility, especially in dense carbons with tortuous structures.

Design implications: prioritize (i) micropore development to maximize capacity, (ii) mesoporous pathways to facilitate diffusion, and (iii) avoidance of excessive tortuosity that limits accessibility. Importantly, BET surface area and pore-size distribution (PSD) data should always be interpreted in the context of soil conditions, since factors such as moisture, DOC, and texture constrain the effective utilization of internal sites in the field.

Surface Functional Groups and Chemical Interactions

Functional groups (O-, N-, P-, S-, and Si-containing moieties) mediate chemisorption via surface complexation, hydrogen bonding, and coordination, thereby shrinking the labile pool that drives exposure (Flórez et al. 2024). Oxygenated groups, such as –COOH, phenolic –OH, and carbonyls, dominate cation binding (J. Chen et al. 2024); N-bearing motifs and other heteroatom sites modulate affinity and electron density. Although typically less abundant, P, S, and Si functionalities also contribute via complexation or by nucleating co-precipitates (D. Luo et al. 2023; M. Luo et al. 2023).

Group abundance originates from feedstock biochemistry and systematically declines with increasing pyrolysis temperature—a trade-off between reactive surface chemistry and carbon stability, as illustrated in Fig. 2 (He et al. 2022). Thus, aligning functional-group composition with soil chemistry and target metal speciation is a key design principle.

(a)

(b)

Fig. 2. Temperature-driven biochar reactive sites and heavy-metal immobilization mechanisms. (Barszcz et al. 2024; Nguyen et al. 2024). (a) Major immobilization pathways of heavy metals by biochar, including precipitation, redox-mediated transformation, surface complexation, electrostatic sorption, hydrogen bonding, π–metal interactions, and pore filling. These pathways are regulated by engineering levers such as alkalinity adjustment, mineral impregnation, redox-active minerals, porosity, surface activation, and surface charge. (b) Pyrolysis temperature drives the evolution of biochar reactive moieties, including the transition from aliphatic to aromatic carbon structures, the transformation of oxygen-containing functional groups, nitrogen species, silicon forms, and mineral phases. Matching these temperature-dependent functional-group and mineral “palettes” to metal speciation and soil chemistry is central to effective biochar design, rather than relying only on bulk O content or BET-derived surface metrics.

Fig. 3. Decision framework linking soil conditions and contamination types to biochar modification strategies

Alkalinity, pH, and Surface Charge (pHₚ_zc)

Soil and biochar pH jointly control metal speciation and sorption thermodynamics. Increasing pH generally lowers the solubility of many cationic metals and promotes hydroxide formation and carbonate-related precipitation, whereas acidification can remobilize carbonate- or hydroxide-associated metals (Kicińska et al. 2022; Z. L. Li et al. 2022). The point of zero charge (pHₚ_zc) frames electrostatics: at pH > pHₚ_zc surfaces are negatively charged and promote uptake of metal cations; at pH < pHₚ_zc positively charged surfaces can retain oxyanions (e.g., HAsO₄²⁻, HCrO₄⁻) (El-Wakil et al. 2022). Representative values in Table S2 show that biochar pH can vary widely, ranging from about 5.26 to 13.15 depending on feedstock and pyrolysis conditions. In general, pH tends to increase with pyrolysis temperature, although substantial variability remains due to feedstock composition and ash content. With increasing pyrolysis temperature, biochar alkalinity tends to rise and pHₚ_zc can shift upward, while cation-exchange capacity often declines due to loss of acidic groups—an important trade-off to report and leverage (Joshi et al. 2023).

Design implications

For acidic soils with cationic metals, alkaline/ash-rich chars raise pH and provide basic exchange sites; for oxyanion risks (e.g., As, Cr(VI)), pH control around pHₚ_zc and provision of specific inner-sphere sites are more decisive than bulk alkalinity alone. Practitioners should balance the benefits of alkalinity against potential losses in cation exchange capacity (CEC) and functionality toward anions.

Engineering Biochar via Physical/chemical Modification: Toolbox, Mechanisms, and Trade-offs

Post-synthesis modification expands the design space of biochar by enabling selective tuning of surface chemistry, porosity, and charge properties beyond what feedstock and pyrolysis alone can achieve (Ravindiran et al. 2024). Strategies include physical and chemical activation, mineral loading, heteroatom enrichment, and biological functionalization. Each method offers distinct advantages but also introduces trade-offs that may constrain field applicability (Wen et al. 2023). A critical evaluation of these approaches is essential to translate laboratory improvements into reliable remediation strategies.

Physical activation (steam/CO₂)

Physical activation enhances surface area and pore volume through controlled gas–solid reactions, typically using steam or CO₂. Steam activation tends to produce higher microporosity, while CO₂ yields a more balanced micro–mesopore network that facilitates diffusion. Reported increases in BET surface area often correlate with improved sorption in batch experiments (Zhang et al. 2021). Under optimized physical activation conditions, BET surface area can exceed 400 m²/g; for example, almond shell biochar reached 436 m²/g (Fan et al. 2022; Table S2).

However, reliance on BET alone can be misleading: larger surface areas do not necessarily translate to improved immobilization in soils, because DOC, colloids, and soil texture can constrain site accessibility (He et al. 2024). Over-activation may collapse micropores and strip oxygen functionalities, weakening chemisorption (Han et al. 2020; Manning et al. 2023). Physical activation is best viewed as a capacity-enhancing baseline that is particularly effective in neutral to slightly alkaline soils with cationic metals. In DOC-rich soils, pore blocking reduces efficacy unless combined with mineral phases to outcompete DOC binding.

Chemical activation (acid/base routes)

Chemical activation modifies the carbon matrix and surface chemistry via acid or alkali treatments. Acid activation (e.g., H₃PO₄, H₂SO₄, HNO₃) removes ash, introduces oxygenated groups, and lowers pHₚ_zc, thereby enhancing inner-sphere complexation with Cd²⁺ and Pb²⁺ (Murtaza et al. 2022). Base activation (e.g., KOH, NaOH) etches microporosity and increases surface area, sometimes boosting sorption capacities by an order of magnitude (Chen et al. 2021).

Nonetheless, acid activation may lead to leaching of phosphate or sulfate, contributing to eutrophication (Cui et al. 2022), while base activation can strip oxygen functionalities and leave residual ions (K⁺, Na⁺) that elevate soil salinity. Many studies emphasize surface area increases without evaluating soil-relevant endpoints (Vijay et al. 2021). Acid routes are better suited for neutral–alkaline soils with cationic metals, whereas base routes are effective in acidic soils but limited against oxyanions. For complex or DOC-rich systems, chemical activation alone is insufficient and should be combined with mineral or heteroatom modification.

Mineral loading and composite formation (Fe/Al/Mn oxides, phosphates, clays)

Incorporating mineral phases introduces high-affinity sorption sites via complexation, ligand exchange, and co-precipitation. Fe/Al oxides enhance retention of oxyanions (As, Cr), while phosphate-rich additives promote cation precipitation (Alam et al. 2020; Zheng et al. 2020). Mn-oxide and MgO composites offer rapid Pb removal with high capacities (Ling et al. 2017), and magnetic composites facilitate recovery and reuse (Song et al. 2021).

Despite their promise, mechanistic attribution is often oversimplified. Reported improvements may reflect unwashed residues rather than true sorption (Liu et al. 2022). Stability is a concern: Fe/Al phases may dissolve under anoxic conditions, and phosphate composites can leach nutrients. Excessive loading may block pores and reduce accessibility. These risks highlight the need for systematic reporting of mineral phase identity, crystallinity, and leaching behavior—data that remain scarce in current literature.

Taken together, mineral loading and composite formation can substantially expand the functional range of biochar for target-specific remediation, particularly in systems requiring oxyanion retention or mineral-assisted precipitation. However, their practical value depends on careful control of loading intensity, phase stability, and leaching risk under soil-relevant conditions.

Heteroatom enrichment (n, s, p motifs)

Doping with nitrogen, sulfur, or phosphorus alters electron density and surface charge, introducing donor sites that enhance metal affinity. N-doping (pyridinic, graphitic N) improves binding with transition metals (Kasera et al. 2022), while sulfur motifs form strong complexes with Pb²⁺ and Hg²⁺ (Yin et al. 2022). P-enrichment promotes precipitation with Cd²⁺ and Zn²⁺ (Peiris et al. 2023).

However, attribution to heteroatoms is often speculative. Many studies fail to distinguish true heteroatom–metal coordination from indirect effects such as pore restructuring or mineral residues (Y. Sun et al. 2024). S- and P-enriched chars may suffer from functional group instability during aging, while N-enrichment may increase NO₃⁻ release. Rigorous mechanistic confirmation using advanced spectroscopy and aging trials is therefore still needed.

Biological modification strategies

Microbial residues and live consortia introduce sorption-active moieties and create micro-environments that complement carbon–mineral scaffolds. Phosphate-solubilizing bacteria (PSB) coupled with biochar enhance Pb immobilization via local phosphate release (H. Chen et al. 2019), while biologically treated residues improve Cd binding efficiency (Tao et al. 2021).

Yet, translation from solution systems to soils remains uncertain. DOC, competing ions, and redox dynamics constrain microbial survival and functionality. Exogenous inocula may disrupt native communities, and extreme pH/salinity limits persistence. Biological modification is best suited for moderate contamination where immobilization and soil–biota recovery are joint goals, but robust field evidence is still limited.

Modification decision map

Figure 3 illustrates a decision framework linking soil–metal scenarios to suitable biochar modifications. The first determinant is soil pH. In acidic soils (pH < 6.5), alkaline or unmodified biochars are effective through pH neutralization and carbonate precipitation (S. Liu et al. 2025). Physical or chemical activation (e.g., steam, alkali) can further increase sorption capacity, but those processes may erode functional groups and generate persistent free radicals (PFRs) or reactive oxygen species (ROS), raising ecological concerns (Z. Zhao et al. 2025). In alkaline soils (pH > 7.5), acidified or Fe/Mn-modified biochars, or low-temperature feedstocks are preferred to avoid further alkalization, though excessive mineral loading may destabilize under reducing conditions or increase nutrient leaching (Lin et al. 2025; L. Liu et al. 2023).

The second decision point is DOC. In high-DOC soils, high-temperature or acid-washed biochar, combined with Fe/Mn or phosphate composites work together to minimize DOC release and enhance co-precipitation (Ahmed and Aidi 2025; T. Sun et al. 2024). Heteroatom doping (N, P, S, Si) improves selectivity but raises questions regarding long-term stability and dopant fate (J. Zhao et al. 2025; Yang et al. 2025). When DOC is low, metal speciation becomes decisive. For cationic metals (Cd, Pb, Zn, Cu, Cr(III)), functionalized biochar or clay/mineral composites enhance immobilization, while anionic metals (As, Sb, Cr(VI)) require Fe/Mn-modified, acidified, or phosphate-loaded biochar (Ahmed and Aidi 2025). Multi-metal contamination favors composite hybrids (Fe–P, clay–biochar) to exploit multiple immobilization pathways (Wang et al. 2025).

Biological modifications provide eco-friendly synergies with microbes or enzymes, but results remain inconsistent in field soils due to competition and abiotic variability (Fakhar et al. 2025).

Despite these advances, key gaps remain. Cross-comparison under standardized matrices and long-term validation are scarce; climate and soil aging can reverse remediation benefits (Dimitriadou et al. 2025). Standardized reporting of pore architecture and functional groups would enable reliable meta-analyses (Schmidt et al. 2021). Integration with AI offers opportunities to optimize multi-parameter designs across soils, feedstocks, and metals, though robustness depends on training data quality (Ge et al. 2025).

Overall, Fig. 3 and Supplementary Table S2 provide a foundation for aligning soil–metal conditions with targeted engineering routes. However, the limitation of single-route optimization becomes especially evident in multi-metal systems. In a Cu-, Zn-, As-, Cd-, and Pb-contaminated soil, nZVI@BC was shown to simultaneously engage multiple mechanisms, including surface adsorption, electrostatic attraction, ion exchange, co-precipitation, oxidation-reduction, and complexation, with competitive and synergistic interactions observed among metals (Song et al. 2022). Likewise, in Cd-, Cr-, and Pb-contaminated soil, a biochar- and bentonite-supported nZVI composite promoted the co-existence of adsorption, reduction, co-precipitation, and complexation during simultaneous immobilization (Jin et al., 2023). These studies support a key inference: under mixed-contaminant conditions, multifunctionality is often observed as an emergent property of composite or modified materials rather than as a fully controlled active-site design. This pattern suggests that, rather than assuming that a single engineered active site can be reliably matched to heterogeneous field conditions, a more practical route may be to combine complementary functionalities—either within one multifunctional biochar or across blended biochar systems—to broaden coverage across cationic and oxyanionic metals.

In practice, such an approach could involve either multifunctional biochars carrying multiple reactive moieties within the same material, or blended amendment systems composed of contrasting biochars or biochar-based composites with different dominant functions, so that more than one immobilization pathway can be deliberately engaged. Function-specific modification illustrates the same principle at a narrower scale. For instance, phosphorus-loaded biochar-assisted phytoremediation was reported to enhance metal immobilization in Cd-, Cr-, and Pb-contaminated soils, mainly through precipitation- and complexation-related processes, while also improving selected soil physicochemical properties (Serrano et al., 2024). Although multifunctional and composite biochars are already well represented in the literature, explicit blended-biochar strategies remain comparatively underexplored. Integrating multiple functionalities within a single biochar may also increase preparation complexity and cost, which may limit routine use but still make such materials attractive for targeted remediation scenarios where broader functional coverage justifies the added engineering burden. A portfolio-style approach may therefore provide a pragmatic route for multi-metal remediation by broadening functional coverage rather than maximizing a single mechanism.

ADDITIONAL ADVANTAGES OF USING BIOCHAR FOR REMEDIATING HEAVY METALS IN SOIL

Crop toxicity due to heavy metals primarily results in reduced yields and the excessive accumulation of these metals. Crops that are subjected to heavy metal contamination may have reduced enzyme activity, deterioration of chlorophyll pigments, and impaired nutrient uptake, among other physiological effects (Hafeez et al. 2023). On the other hand, heavy metal accumulation in crops can transfer through the food chain to affect human health (Mwelwa et al. 2023). This section reviews studies focused on increasing crop yields and reducing the accumulation of heavy metals.

Increasing Crop Yields

Biochar may support crop productivity in heavy-metal-contaminated soils by improving soil physical and chemical conditions and thereby alleviating stress at the root–soil interface. However, these benefits are not uniform and should not be generalized across all biochars. A more meaningful interpretation requires comparison across feedstock type, application rate, and thermochemical history, because these factors often influence whether biochar improves or constrains soil functioning.

A first distinction concerns feedstock type. In many cases, lignocellulosic biochars, such as those derived from crop residues, are associated with improvements in pore architecture, aggregate stability, and water retention, which may indirectly support crop growth. For example, rice-straw biochar was reported to improve bulk density, porosity, aggregate stability, and soil hydraulic pathways, thereby contributing to a more favorable air–water balance in soil (Wu et al. 2023; Ramezanzadeh et al. 2023). By contrast, waste-derived biochars, such as sludge biochar, may also improve physical properties, but their effects often appear to be more strongly influenced by ash content, mineral composition, and greater feedstock heterogeneity. Xin et al. (2023) found that sludge biochar altered particle-size distribution, decreased bulk density, and increased total porosity by 1.2 to 11.5%. In addition, waste-derived biochars often contain higher initial nutrient and ash contents, and ion-exchange processes during metal sorption may release certain nutrient ions into the soil solution. Under some conditions, this mechanism may contribute to improved nutrient availability and crop performance, although the magnitude of this effect remains strongly context-dependent. Taken together, the available evidence suggests that lignocellulosic biochars are often linked to structural and hydraulic improvements, whereas waste-derived biochars may exert stronger mineral-related or liming effects, albeit with greater variability across feedstocks.

A second determinant is application level. At low-to-moderate levels, biochar often improves plant-available water, porosity, and root-zone conditions, which may contribute to better crop performance. For instance, all treatments considered by Šurda et al. (2024) significantly increased plant-accessible water content and porosity in sandy soil, while X. Chen et al. (2023) showed that a 2% application rate exerted a more pronounced positive effect on water retention in medium-grained soils than in coarse-grained soils. However, higher application levels do not necessarily yield proportionally greater benefits. Z. Chen et al. (2024) observed that when biochar application exceeded 4%, saturated permeability declined by approximately one order of magnitude. The effect was attributed to pore filling within clay intra-aggregate spaces during repeated freeze–thaw cycles. These findings suggest that dosage may act as a threshold-dependent factor: moderate rates are often beneficial, whereas excessive rates may impair permeability or aeration under certain soil conditions.

A third consideration is thermochemical history, especially pyrolysis temperature, although the evidence in this subsection is less systematically reported than feedstock and dosage effects. In general, pyrolysis temperature influences the balance between short-term reactivity and longer-term persistence, and may therefore affect nutrient dynamics, water relations, and the durability of agronomic benefits (He et al. 2022; Joshi et al. 2023). However, the available literature does not support a universal ranking of low- versus high-temperature biochars for crop-yield improvement, because temperature effects are typically mediated by feedstock type, particle size, soil texture, and application rate. Pyrolysis temperature should therefore be interpreted as a modifying factor rather than as an independent predictor of agronomic performance.

Beyond these broad comparisons, the evidence on soil hydraulic responses remains mixed. Biochar may enhance water retention and hydraulic conductivity by increasing mesoporosity and improving pore connectivity, but under other conditions it may reduce permeability by clogging fine pores or altering the packing arrangement between soil particles and biochar particles. Faloye et al. (2024), for example, reported significant improvements in saturated hydraulic conductivity in biochar-amended soils, attributing this to increased mesopore development. In contrast, Z. Chen et al. (2024) found decreased permeability under specific high-rate application and freeze–thaw conditions. Similar variability has been reported for shrinkage, cracking, and air permeability, all of which appear to depend on the interaction between soil texture, biochar particle size, pore structure, and amendment rate.

Overall, the yield-related benefits of biochar are best understood as context-dependent co-benefits rather than universal outcomes. Lignocellulosic biochars are often associated with physical improvement and water regulation; waste-derived biochars may provide stronger chemical or liming effects but with greater variability; moderate dosages are generally more reliable than excessive applications; and pyrolysis temperature appears to influence the balance between reactivity and persistence. Accordingly, improvements in crop yield should be interpreted through a comparative framework that considers feedstock origin, dose, thermochemical history, and soil constraints together, rather than through isolated case studies.

Based on such findings, the main advantages of biochar in this context include improved soil physical functioning, enhanced water regulation, and potential support for crop performance, whereas the main limitations include variability across feedstocks, threshold-dependent dosage effects, and inconsistent hydraulic responses under different soil conditions.

Reducing the Accumulation of Heavy Metals

Some studies attribute the reduction in heavy metal accumulation in crops following biochar application to its ability to decrease the bioavailability of heavy metals in soil (Fig. 4).

Fig. 4. Proposed mechanisms for heavy-metal adsorption and immobilization by biochar modified from (W.-H. Chen et al. (2022) and Wang et al. (2019) ions between pyrolysis temperature, elemental composition, and resulting reactive moieties of biochar; implications for pollutant removal mechanisms. (Adapted based on the scientific framework reported by He et al. 2022).

Notably, plant uptake does not always scale linearly with bulk-soil bioavailability indices, because rhizosphere processes and physiological barriers can dominate under certain conditions. In this context, the impact of biochar on crop metal accumulation should be understood as a coupled soil-rhizosphere-plant process rather than as a simple consequence of reduced extractable metal pools alone. Other studies further suggest that biochar may influence plant physiological and biochemical responses under metal stress, thereby contributing to lower uptake and accumulation. Haider et al. (2022) summarize that biochar application can regulate stress-induced antioxidant responses in plants. Biochar application may also affect membrane-associated transport and cell permeability. Biochar also enhances ATPase activity, which is linked to the synthesis of soluble proteins and nucleic acids, which can aid in metal contamination control. Additionally, biochar reduces ROS formation under metal stress, thereby alleviating oxidative damage in polluted soils. Through these combined effects, biochar may support metabolic and enzymatic activity and reduce contaminant transfer into plant tissues. Overall, many studies report simultaneous yield benefits and reduced metal accumulation following biochar amendment; however, the magnitude and consistency of these responses vary with soil constraints, metal chemistry, biochar properties, and management context (Haider et al. 2022; Y. H. Liu et al. 2023).

While these agronomic co-benefits motivate interest in biochar as a multifunctional amendment, farm-scale remediation decisions are typically made in the context of alternative organic inputs and practical trade-offs.

A complementary perspective is to contrast engineered biochar with more labile organic amendments commonly used in farming systems (e.g., compost or manure). Biochar, as a thermochemically transformed and increasingly “engineerable” carbon material, can be tailored toward sorption, complexation, and precipitation functions through feedstock choice, pyrolysis conditions, and post-modification. A practical advantage of this functional tunability is that targeted functions can be deliberately strengthened to improve reproducibility, whereas organic amendments often exhibit greater variability in composition and reactivity across feedstocks and composting maturity. At the same time, engineered biochar should not be treated as synonymous with activated carbon, although partial overlap can occur. Biochar remains defined by the thermochemical conversion of biomass and its broader agronomic or environmental functional context, whereas activated carbon is more commonly regarded as a more intensively processed sorbent optimized primarily for high surface area and adsorption performance (Subramanian et al. 2025). Some strongly activated or post-modified biochars may enter an overlapping functional space, but many engineered biochars retain design goals extending beyond sorption alone. While biochar may provide ancillary agronomic co-benefits (e.g., liming), its key engineering distinction lies in its relative persistence, which supports longer-lived immobilization when governing mechanisms remain stable. In contrast, conventional organic amendments contribute more directly to nutrient supply and stimulation of biological activity, but their impacts on metal immobilization are often more dynamic; increases in dissolved organic carbon (DOC) and competing ligands may either attenuate or enhance metal mobility depending on soil redox–pH trajectories (Xu et al. 2025). Nevertheless, organic amendments can serve as practical, low-intensity options for mildly contaminated soils; for example, compost has been reported to reduce Cd phytotoxicity while improving wheat growth in specific case studies (Ahmad et al. 2025).

Taken together, these contrasts underscore that remediation in agricultural soils is inherently multi-objective, requiring a balance between food-safety risk reduction and fertility management. Outcomes are strongly contingent on baseline soil constraints, contamination severity, and management context. Accordingly, strategy selection benefits from calibration to local conditions and may include integrated approaches that combine sorption-oriented biochars with nutrient-oriented organic inputs where appropriate. In this context, the main advantages of biochar are its relative persistence, engineerability, and potential to simultaneously reduce crop metal uptake and support agronomic performance, whereas its main limitations lie in context-dependent effectiveness, variability across feedstocks and soils, and the possibility of unintended trade-offs under field conditions. However, regardless of the pathway chosen, introducing amendments—especially engineered materials—into soil can generate environmental trade-offs and unintended consequences, which are critically evaluated in the following section.

ENVIRONMENTAL AND OPERATIONAL RISKS OF BIOCHAR APPLICATION

Biochar is widely promoted as a cost-effective and multifunctional amendment for agricultural soils. However, its production and application are not without environmental and operational risks (Long et al. 2024; Yao et al. 2024). A critical assessment of these risks—ranging from pollutant formation during pyrolysis to unintended ecological perturbations in field soils—is essential for translating laboratory promise into safe and scalable remediation practices.

Risks during Biochar Preparation

Thermochemical conversion of biomass can generate polycyclic aromatic hydrocarbons (PAHs) and dioxins, both of which pose environmental and health concerns (Hale et al. 2012). PAHs are formed during pyrolysis and related thermochemical processes, with concentrations strongly influenced by feedstock type, production method, residence time, and temperature. Reported total PAH concentrations in biochars vary widely; for example, Hale et al. (2012) reported values of 0.07 to 3.27 μg g^-1 for slow-pyrolysis biochars, compared with 0.3 μg g^-1 for a fast-pyrolysis biochar and 45 μg g^-1 for a gasification biochar. In that study, the highest PAH concentrations in slow-pyrolysis materials were generally observed in biochars produced in the 350 to 550 °C range, while higher pyrolysis temperatures often resulted in lower total PAH concentrations. Excessive activation or poorly controlled processing can also favor PAH accumulation (Weidemann et al. 2018).

Dioxin formation is mainly associated with feedstocks rich in chlorine and with low-temperature processes such as hydrothermal carbonization. Studies indicate that chars produced at 200 to 300 °C pose the highest risk of generating chlorinated congeners with elevated toxicity (Sobol et al. 2023). Regulatory frameworks such as the European Biochar Certificate (EBC) set a limit for PCDD/F (dioxins and furans) at 20 ng/kg dry matter (I-TEQ, EBC guidelines) (Grafmüller et al. 2024). However, compliance with this threshold and transparent reporting of dioxin/furan levels in biochar products is still uneven across studies.

Overall, feedstock choice, pre-treatment, and pyrolysis control are decisive for minimizing contaminant formation, yet these aspects are insufficiently standardized across studies.

Risks during Biochar Application

Although biochar is widely promoted for heavy-metal immobilization and soil remediation, its application stage poses several risks that depend heavily on soil chemistry, biochar feedstock and production conditions, and field management practices. Key risks include:

Secondary release of heavy metals

Biochar can act as a transient sink rather than a permanent fix: under acidic pH or fluctuating redox conditions, adsorbed metals (e.g., Cd, Pb, Zn) may re-mobilize. At the same time, such re-mobilization does not necessarily imply irreversible loss of retention, as released metal ions may be re-adsorbed by biochar or associated mineral phases as the system continues to equilibrate, provided that sufficient sorption capacity remains and local chemical conditions remain favorable. A column experiment using maize straw–derived biochar pyrolyzed at 600 °C showed increased mobility of Cu, Pb, Zn, and Cd in a highly acidic soil (pH ~4 to 5) when large biochar doses were applied, while soils with moderate to high pH showed much reduced remobilization (Jia et al. 2021). Additionally, a 3-year field experiment in paddy soil revealed that biochar amendment significantly decreased exchangeable Cd fractions, but the study also suggested that biochar aging and soil buffering capacity are crucial for ensuring long-term immobilization (Sun et al. 2023).

A management lesson provided by the forementioned studies is to select feedstocks with low inherent heavy-metal content; combine biochar with stabilizing mineral phases (e.g., iron or manganese oxides), especially in acid or redox-variant soils; and implement leaching tests under site-relevant conditions and monitor over multiple seasons.

PFRs and redox toxicity

Biochar, especially from certain biomass sources and pyrolysis conditions, frequently contains environmentally persistent free radicals. These radicals can catalyze ROS formation, with potential to damage microbial cells or soil fauna. Alfei and Pandoli (2024) reported that biochars commonly contain abundant carbon- and oxygen-centered radicals, whose relative abundance varies with feedstock type and pyrolysis temperature. However, field data quantifying ecological impacts of PFRs (e.g., on microbial oxidative stress, plant health) remain sparse and inconsistent.

The management lesson in this case is to employ harmonized protocols for quantifying PFR concentration; avoid pyrolysis conditions that maximize radical formation without controlling quenching; investigate aging or post-treatment (e.g., natural oxidation, steam, or light exposure) to reduce reactive radical concentrations before widespread field application.

Soil microbiome perturbations

Biochar modifies soil microhabitats through changes in pH, adsorption of inhibitory molecules, or altering moisture and substrate availability. While many lab-scale, short-term studies report increased microbial biomass or enzyme activities, longer-term or field scale work reveals shifts in community composition and function that may be detrimental (e.g., nitrification/denitrification reductions). Bolan et al. (2024) reported positive or negative effects, depending on biochar rate, feedstock, and soil type.

The management lesson here is to use moderate biochar amendment amounts; co-apply buffering amendments (organic matter, compost) to offset potential adverse pH or substrate effects; and include multi-season microbial community and functional enzyme monitoring in field trials to detect delayed or cumulative effects.

Agrochemical interactions (pesticide/herbicide interception)

Biochar’s sorptive properties can reduce off-site losses of agrochemicals , but they may also decrease pesticide bioavailability and thereby compromise pesticide efficacy under field conditions. A global meta-analysis (J. Wang et al. 2024) covering 58 studies (386 observations) found that high-adsorption biochars generally reduced pesticide residues in organisms (plants, earthworms) by 66%, but they had no consistent effect on soil pesticide concentrations. Effects depended strongly on soil properties (e.g., organic matter, texture) and biochar characteristics. Another study with rice straw biochar showed that atrazine, simazine, and prometryn had decreased half-lives in biochar-amended soils, but adsorption also competes with microbial degradation, meaning that timing and dose remain important (Singh et al. 2024).

Taken together, these findings indicate that agrochemical interactions should be considered as a practical management risk of biochar application, particularly where reduced pesticide efficacy may offset other agronomic benefits. Accordingly, biochar-pesticide compatibility should be tested under field conditions, with attention to application timing, dose, environmental fate, and crop protection outcomes.

Intrinsic contaminants in certain biochars

Biochars derived from manures or wastes often have higher ash content and background heavy-metal burdens. For example, in the three-year field trial in paddy soil, biochar application reduced Cd in rice organ uptake, but the study also showed that some Cd remained in more mobile soil fractions, and immobilization efficiency declined over time with aging and under lower biochar dosages (Sun et al. 2023)

The management lesson to be shared from the reported work is to require feedstock screening for metal content, ash composition; to certify batches; to use higher pyrolysis temperature or modifications (washing, mineral amendment) to immobilize resident contaminants; and to track mobile and bioavailable metal fractions, not only total metal content.

Summary

Collectively, these findings underscore that the efficacy of biochar depends not just on immediate sorption or immobilization but on long-term stability, soil buffering, and management under realistic field conditions. Risks may be magnified in acidic, redox-variable, high-DOC, or low-buffer soils. Regulatory or certification frameworks (e.g., European Biochar Certificate) should increasingly require performance-based monitoring (leachability, PFR reactivity, pesticide interception effects, microbial function over time) rather than just compositional limits.

These limitations highlight the importance of moving beyond biochar as a stand-alone amendment and exploring integrated strategies with biological and agronomic partners, which are discussed in the next section.

EMERGING APPLICATIONS: INTEGRATING BIOCHAR WITH BIOLOGICAL AND AGRONOMIC APPROACHES

Biochar is widely applied as a low-cost and environmentally friendly amendment for pollutant removal and soil remediation (Gong et al. 2022). As has been detailed, its primary function in contaminated soils is stabilization rather than degradation—shrinking labile metal pools via sorption, surface/co-precipitation, and pH/charge effects (Mukherjee et al. 2022). Building on these mechanisms, integrating biochar with microorganisms, plants, or soil fauna can enhance remediation efficiency through synergistic pathways (Xiang et al. 2022). An at-a-glance summary of representative pairings, mechanisms, and suitable soil–metal contexts is provided in Table S3.

Biochar Alone for Heavy Metal Immobilization

Biochar used as a stand-alone amendment has been widely investigated for heavy-metal immobilization, but its efficacy remains highly soil-specific (Guo et al. 2020). As discussed earlier, the main mechanisms include sorption, surface or co-precipitation, and pH buffering, although their relative importance varies with metal speciation, soil chemistry, and biochar properties. Case studies nonetheless confirm that stand-alone biochar can be effective under suitable conditions. For example, Li et al. (2018) reported nearly 98% removal of Cr(VI) using H₃PO₄-activated corncob biochar. In such applications, performance is shaped largely by the interaction between biochar’s thermochemical history (e.g., pyrolysis temperature) and feedstock origin, with higher-temperature chars often showing greater capacity for redox-mediated Cr(VI) transformation.

This redox dimension is particularly relevant for chromium, because biochar has been reported to participate in the transformation of highly toxic Cr(VI) to the less toxic and more readily immobilized Cr(III). Such participation may occur through electron-shuttling roles in microbially mediated systems, as well as through electron-donating or electron-mediating functions under abiotic conditions (Ren et al. 2023; Xu et al. 2021). In biochar-based composites, this transformation may be further enhanced, with biochar functioning mainly as an electron mediator while redox-active Fe species serve as the principal electron donors (Dai et al. 2025). Even so, converging evidence indicates that mineral-modified or composite biochar generally outperforms stand-alone materials in both capacity and stability (L. Sun et al. 2022; Tan et al. 2023).

The practical value of stand-alone biochar lies in its relative simplicity, lower modification burden, and ability to provide baseline immobilization under favorable soil conditions. However, its main constraint is the limited reliability of performance under multi-metal contamination, redox-variable environments, or otherwise heterogeneous field conditions. These limitations motivate the exploration of integrated strategies (see Table S3), which are discussed in the following subsections. Accordingly, long-term field validation remains necessary to determine when stand-alone biochar is sufficient and when more engineered or integrated systems are required.

Biochar and Microorganisms Synergistic Remediation

Compared with direct microbial loading discussed in the section on biological modification strategies, where biochar is pre-loaded or surface-modified with biomass, synergistic remediation systems emphasize the in-situ co-functioning of biochar and soil microbiota. Here, biochar acts less as a passive carrier and more as a catalyst for functional hotspots in which physicochemical stabilization and biological transformation converge. Within biochar pores and on its surfaces, microorganisms accelerate biomineralization (via localized carbonate/phosphate release), drive redox conversions (e.g., Cr(VI) → Cr(III)), and secrete extracellular polymeric substances (EPS) that complex metals into less labile forms; co-localization further creates co-sorption hotspots and shifts community composition toward metal-tolerant guilds (Garg et al. 2025; Manikandan et al. 2023; Shan et al. 2021) (Fig. 5).

Different biochar–microbe interaction modes can be compared in terms of their dominant remediation functions and practical limitations. In the simplest systems, co-application with beneficial microbial inoculants, such as phosphate-solubilizing bacteria (PSB), may enhance immobilization beyond either component used alone. For example, PSB co-applied with biochar improved Pb and Cd immobilization more effectively than single-component treatments (Lai et al. 2022). A second category involves functionally modified biochars that appear to favor beneficial microbial guilds or sustain microbial activity under stress conditions. Phosphorus-modified biochar, for instance, was reported to promote beneficial microbial communities and maintain remediation performance under contamination stress (Y. C. Wang et al. 2024). A third category includes redox-active or composite systems, in which biochar and microorganisms may jointly facilitate detoxification through coupled electron-transfer processes and redox transformation pathways (Manikandan et al. 2023). Taken together, these studies suggest that biochar–microbe systems may be most effective when they combine physicochemical immobilization with biologically mediated transformation, rather than relying on either pathway alone.

Fig. 5. Mechanisms of heavy-metal bioremediation by Bacillus species immobilized on biochar (Manikandan et al. 2023, under CC BY 4.0 license)

At the same time, these advantages should not be overgeneralized. The persistence of exogenous inocula under fluctuating pH, moisture, and redox conditions remains uncertain, and competition with native microbial communities may offset intended benefits. In addition, positive responses observed in controlled systems do not always translate directly to heterogeneous field soils, where contamination history, soil texture, and DOC dynamics may simultaneously affect microbial establishment and metal mobility. In particular, DOC may play a dual role by supporting microbial activity as a carbon source while also enhancing metal mobility through complexation under some conditions. Thus, the apparent success of biochar–microbe consortia often depends not only on amendment design, but also on ecological compatibility and environmental stability. Co-application systems mainly strengthen immobilization efficiency. Functionally modified biochars mainly support microbial persistence under stress, whereas redox-active composites are more relevant where detoxification through electron-transfer pathways is required.

Overall, biochar–microorganism synergistic remediation is promising because it may couple immobilization, transformation, and ecological recovery within the same system. However, the current evidence remains stronger for proof-of-concept performance than for field reliability. Future studies should therefore prioritize multi-season validation, direct comparison with non-biological amendment systems, and joint monitoring of microbial community dynamics and labile metal pools. Such work will be essential to determine when microbial synergy provides a robust remediation gain and when it remains conditional on site-specific constraints.

Biochar and Phytoremediation Integrated Techniques

Beyond the general agronomic benefits of biochar summarized earlier, its integration into phytoremediation systems specifically enhances remediation efficiency by modifying the rhizosphere environment—reducing the labile fraction of metals, stimulating root-associated enzymes, and supporting plant stress tolerance. For example, P. Sun et al. (2022) observed that combined biochar–plant treatments increased Pb immobilization by 12 to 18% compared with either treatment alone. Similarly, ferrate-modified biochar improved ryegrass tolerance in petroleum–zinc co-contaminated soils, enhancing hydrocarbon degradation while limiting Zn accumulation (M. R. Zhang et al. 2024). Mushroom-derived biochar co-applied with Medicago sativa also reduced bioavailable Cd by 44% in multi-metal soils while sustaining plant establishment (A. D. Wang et al. 2023).

Overall, biochar–plant combinations exhibit strong synergistic effects, enhancing both contaminant immobilization and plant productivity. However, most evidence stems from short-term pot experiments, and the efficiency of these systems under field-scale, multi-season conditions remains underexplored. Future work should clarify crop-specific responses, long-term ecological trade-offs, and the economic feasibility of large-scale deployment.

Biochar–soil Fauna synergistic Remediation

Soil fauna, particularly earthworms, can facilitate heavy metal remediation through bioaccumulation, biotransformation, stimulation of microbial activity, and enzyme release (C. Zhang et al. 2024). When combined with biochar, these processes can be synergistic. Biochar additions have been reported to elevate earthworm metallothionein levels, enhancing Cd tolerance (Luo et al. 2024), while vermicomposting systems integrating biochar produce “vermibiochar,” which is enriched with enzymes and organic ligands that further stabilize metals (Rehman et al. 2023). Such combinations can substantially reduce the bioavailability of toxic metals. For instance, Alsamhary (2023) documented a 60% reduction in Cd bioavailability when biochar was co-applied with Eisenia fetida.

Nevertheless, findings remain inconsistent. Certain studies suggest that some earthworm species may actually increase heavy metal mobility by stimulating nitrification and lowering soil pH, thereby undermining biochar’s immobilization capacity (J. Wang et al. 2023). These contrasting outcomes indicate that biochar–earthworm systems are context-dependent, influenced by factors such as soil chemistry, biochar type, and species-specific earthworm behavior.

In summary, biochar–animal synergistic approaches represent an innovative frontier in remediation research, offering potential co-benefits for soil fertility and structure. However, their mechanisms remain poorly understood, and evidence from long-term field trials is scarce. Future studies should prioritize mechanistic elucidation, species selection, and ecological risk assessments before recommending widespread adoption.

ARTIFICIAL INTELLIGENCE FOR BIOCHAR-BASED HEAVY-METAL REMEDIATION IN SOILS: FROM MATERIAL DESIGN TO DECISION SUPPORT

Artificial intelligence (AI) is increasingly being explored as a decision-support tool for addressing the complexity of soil-biochar-metal interactions in agricultural remediation systems (Balakumar et al. 2025). In contrast to conventional empirical or regression-based approaches, AI enables the integration of heterogeneous datasets, supports efficient parameter optimization, and delivers robust predictive capabilities for highly nonlinear systems (Ahmed and Aidi 2025). Within the domain of biochar-based soil remediation, AI applications are converging along two principal axes: (i) the rational design and process optimization of biochar materials (Zhou et al. 2024), and (ii) the prediction, monitoring, and adaptive control of remediation performance (Li et al. 2024). This integration is conceptually illustrated in Fig. 6, while representative AI/machine learning (ML) implementations in biochar research are summarized in Table S4. Importantly, although AI offers data-driven accuracy and risk minimization, its practical deployment in biochar remediation remains at an early stage, highlighting the need for systematic frameworks and field-level validation (Guo et al. 2025).

Fig. 6. Workflow of AI-assisted modeling for biochar-based heavy-metal remediation in soil

AI-driven Biochar Design and Process Parameter Optimization

Optimizing pyrolysis conditions to produce biochar with tailored physicochemical properties—such as yield, specific surface area (SSA), carbon content, and pore architecture—remains a critical challenge owing to the complex, nonlinear interplay between feedstock composition and pyrolysis parameters (e.g., temperature, residence time, heating rate) (M. Liu et al. 2025). Traditional trial-and-error approaches are not only labor-intensive and costly but also prone to local optima, which can limit generalizability and scalability (M.-W. Chen et al. 2023). This bottleneck has constrained the wider deployment of biochar in environmental and agricultural applications. In response, AI-driven methodologies are reshaping this paradigm by systematically exploring high-dimensional design spaces, uncovering hidden interactions, and enabling rapid in silico screening (Jiang et al. 2025). These advances can accelerate discovery, reduce experimental overhead, and lay the foundation for predictive, knowledge-guided biochar engineering.

Across recent studies, tree-based ensembles such as Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Gradient Boosting Decision Trees (GBDT), as well as neural networks (NNs), generally outperform linear baselines for predicting yield, SSA, and composition from proximate/ultimate analyses, lignocellulosic features, and reactor parameters. For instance, Uppalapati et al. (2025) found that XGBoost attained a test R² close to 0.89, which was markedly higher than the 0.44 obtained by the linear model Lasso Regression. Beyond point prediction, model interpretability has increasingly become standard practice because it enables the identification of key variables, supports more efficient optimization of biochar design and application strategies, and helps align data-driven findings with plausible physicochemical explanations rather than leaving models as opaque black boxes (Freiesleben et al. 2024; Murdoch et al. 2019). Among various approaches, SHapley Additive exPlanations (SHAP) is widely used, as it not only ranks features by importance but also characterizes how changes in feature values shift model outputs, thereby revealing structured patterns in the available data (Lundberg and Lee 2017). In practice, SHAP values, partial-dependence profiles, and permutation importance frequently highlight pyrolysis temperature, ash content, elemental composition, and residence time as influential drivers of predicted yield (Uppalapati et al. 2025), SSA, and functional-group-related indicators—patterns that are broadly consistent with established pyrolysis chemistry.

Multi-objective optimization is increasingly used to navigate trade-offs among yield, SSA, energy input, and environmental performance. For instance, yield-and-SSA joint modeling with ensembles (Hai et al. 2023) and SSA–Total Pore Volume (TPV) modeling with gradient boosting (Li et al. 2023) exemplifies data-driven design approaches for pore-structure engineering. Collectively, the studies summarized in Table S4 suggest several recurring contributions of AI/ML to biochar research. Across diverse datasets, nonlinear models such as Random Forest, XGBoost, and neural networks often outperform linear baselines in predicting biochar yield, SSA, and composition, indicating an improved ability to capture complex interactions among feedstock composition, pyrolysis conditions, and (where included) soil-context variables. Despite heterogeneity in experimental sources, these works converge on a broadly similar set of predictive inputs—proximate and ultimate composition, lignocellulosic fractions, and reactor parameters—suggesting the emergence of a core variable framework for biochar modeling. Furthermore, many studies emphasize not only predictive accuracy but also interpretability, increasingly adopting SHAP and related techniques to identify dominant predictors and to contextualize model behavior in ways that are consistent with known pyrolysis mechanisms. Taken together, these contributions position AI/ML as a promising and knowledge-oriented toolset for guiding biochar design and informing soil remediation strategies.

AI-assisted Prediction and Control of Heavy-Metal Immobilization

The efficiency of biochar-mediated immobilization is difficult to predict because of nonlinear interactions among biochar properties, soil chemistry, metal speciation, and application strategies (Zand and Abyaneh 2020). Such complexity introduces uncertainty that may cause environmental risks, for example, ineffective immobilization or unintended remobilization (Dong et al. 2025). To address these challenges, AI models—including RF, GBDT, Support Vector Regression (SVR), ANN, and emerging deep or hybrid architectures—are increasingly employed to integrate diverse covariates and provide data-driven forecasts of immobilization outcomes across metals and soil contexts.

Studies summarized in Table S4 suggest that models emphasizing controllable variables (biochar traits and application conditions) typically explain more variance than those centered primarily on intrinsic metal properties. Random-forest frameworks trained on consolidated soil datasets (Guo et al. 2023; Palansooriya et al. 2022) report strong cross-validated performance and commonly use SHAP and Partial Dependence Plots (PDP) to identify leverage points for design-for-remediation. Extensions to soil–plant systems (Li et al. 2024) further indicate promise for predicting crop uptake and bioaccumulation (BAF), although available training data remain dominated by pot experiments and short-term studies. Table S4 provides representative sample sizes, performance metrics, and validation strategies across these implementations.

Importantly, moving from prediction toward practical monitoring and control will require models that remain reliable under field heterogeneity and changing conditions and that can support operational decisions (e.g., amendment selection and dosing) with transparent uncertainty. AI has also been explored for predicting broader soil-process responses to biochar amendment, including greenhouse-gas emissions such as N₂O, showing that its utility may extend beyond remediation efficiency alone (Wang et al. 2024). Although such applications fall outside the core focus of heavy-metal immobilization, they help situate AI-assisted remediation within the wider context of biochar-soil interactions. These considerations motivate the critical challenges discussed next, including data bias, leakage-safe validation, and lab-to-field transferability.

Critical Challenges: Data Bias, Model Reliability, and Validation Gaps

While AI offers substantial potential for biochar remediation research, translation from modeling to agronomic decision support remains constrained by three recurring bottlenecks.

Data limitations constitute a structural constraint on AI credibility in biochar-based remediation. Recent methodological analyses have highlighted that reported accuracy can be systematically inflated in small-n settings, consistent with the higher susceptibility of small datasets—typical of current biochar research—to overfitting and optimistic performance estimates (Vabalas et al. 2019). Literature-derived datasets are also likely enriched for “successful” outcomes, whereas null or unfavorable results are less frequently reported (publication bias), which may lead training data to over-represent best-performing scenarios and under-sample boundary conditions and failure modes critical for risk-aware deployment (Saidi et al. 2025). Robustness is further weakened by heterogeneity and incompleteness: key descriptors (e.g., pore-volume distributions, mineral phases, surface functional group metrics) are often missing or measured inconsistently across studies, introducing measurement bias and forcing imputation or feature exclusion. Finally, dataset imbalance (over-representation of particular feedstocks or soil types) and inconsistent outcome definitions (different extraction protocols) create domain-shift and label-inconsistency risks when models are applied beyond their training distribution (Miftahushudur et al. 2025). Together, these data constraints place a practical ceiling on model reliability and motivate the need for stringent validation and uncertainty-aware reporting.

Beyond data quality, methodological pitfalls in model development pose major risks to generalizability. A pervasive issue is study-structure leakage, where random splits place samples derived from the same study or experimental series (sharing feedstock, soils, treatment settings, and measurement protocols) into both training and test sets (M.-W. Chen et al. 2023). This can inflate apparent performance by evaluating interpolation within familiar experimental regimes rather than true transferability. The problem may be compounded by adaptive data analysis, in which a fixed test set is repeatedly leveraged—explicitly or implicitly—during model selection and hyperparameter tuning, yielding overly optimistic estimates and overfitting to dataset-specific idiosyncrasies (Saidi et al. 2025). Accordingly, leakage-safe evaluation (e.g., grouped/blocked cross-validation such as leave-one-study/site/feedstock-out, and external tests where available) is essential for credible claims of generalization. Moreover, interpretability tools such as SHAP require cautious application. Attributions are model- and feature-set-dependent: different choices of input representation (e.g., inclusion/exclusion of correlated covariates, feature engineering, encoding, and preprocessing) can lead to substantially divergent importance rankings even on the same underlying dataset. Therefore, SHAP results should be interpreted primarily within the context of a specific modeling pipeline rather than as universally comparable “mechanistic truth,” and cross-model comparisons should be grounded in harmonized feature sets and leakage-safe validation designs (Freiesleben et al. 2024; Murdoch et al. 2019).

Finally, a critical lab-to-field generalization gap persists. Much of the available training evidence is derived from short-term, laboratory-scale incubation studies, often using homogenized and/or spiked soils (Vijay et al. 2021). Such datasets do not fully capture field heterogeneity and, crucially, long-term biochar aging processes (e.g., surface oxidation, mineral deposition, and pore clogging). Consequently, AI models trained predominantly on short-term “fresh-biochar” conditions may overestimate remediation performance when transferred to aged biochars or long-contaminated field soils (Cui et al. 2024). This gap highlights that data representativeness can be as important as algorithmic sophistication; future research should prioritize integrating field-scale and long-term datasets, and where appropriate, adopt transfer learning or domain-adaptation strategies to calibrate lab-trained models to field realities. In parallel, decision-grade deployment requires uncertainty-aware outputs and explicit statements of domain applicability when extrapolating across soils, climates, and aging states.

Collectively, these bottlenecks imply that AI in biochar remediation is currently best positioned as bias-aware, uncertainty-calibrated decision support—rather than a universal predictor—until leakage-safe validation and field-scale evidence have become more mature.

CONCLUSION AND OUTLOOK

Conclusions

Biochar has evolved from a simple soil amendment to an increasingly engineerable material platform for environmental remediation. This review has systematically evaluated the state of the art in biochar engineering, highlighting that heavy-metal immobilization is often shaped by a coupled interplay among pore architecture, surface functional groups, and redox-active components. In this review these pathways have been consolidated into a “mechanism–lever–endpoint” logic (Fig. 3), bridging mechanistic hypotheses with practical design and synthesis choices under soil–metal constraints.

However, the inherent complexity of soil–biochar–metal interactions increasingly limits trial-and-error optimization and encourages “black-box” deployment—both in material selection (e.g., overlooking long-term risks such as remobilization) and in model development (e.g., overconfident performance claims under leakage-prone validation). In this context, AI can be a powerful decision-support layer for navigating high-dimensional design spaces and for prioritizing candidate strategies that merit field testing. Yet, it is important to emphasize that neither high sorption capacity nor high predictive accuracy is sufficient for decision-grade use. Risk-aware deployment therefore benefits from a dual-validation mindset: validating materials for durability and ecological safety over time, and validating models for leakage-safe transferability with uncertainty-aware reporting.

Outlook

To transform biochar-based remediation from an empirical practice to a precision science, future research must address critical gaps through a coordinated paradigm shift. Future research should prioritize the following areas:

Standardization for data-driven discovery

The current fragmentation of data hinders the application of advanced analytics. Establishing universal reporting standards for pyrolysis conditions, feedstock properties, and functional parameters is a prerequisite. High-quality, standardized datasets are essential to unlock the full potential of Machine Learning (ML) meta-analyses, ensuring that predictive models are robust and transferable across different soil types.

From soil immobilization to food safety prediction

Current research predominantly focuses on reducing the extractable fraction of metals in soil (e.g., TCLP/DTPA methods). However, reduced bioavailability in soil does not always linearly correlate with reduced accumulation in crop tissues due to the complexity of the soil-root-shoot barrier. A critical future direction is to shift the prediction target from “soil availability” to “edible tissue accumulation.” Integrating AI models with plant physiology data will make it possible to predict heavy metal concentrations in grains or fruits directly. This shift is essential for bridging the gap between geochemical remediation and actual human health risk assessment.

Embedding computational toxicology for “safe-by-design” composites

While modifying biochar with nanomaterials or metal oxides enhances efficiency, it may inadvertently increase cytotoxicity or ecological risks. Future computational frameworks must integrate computational toxicology. By incorporating molecular descriptors of loaded materials into predictive ML models, researchers can achieve multi-objective optimization. This may involve simultaneously maximizing heavy metal immobilization while minimizing the intrinsic toxicity of the engineered biochar. Such a “Safe-by-Design” approach prevents the “regrettable substitution” scenario where a remediation material becomes a new pollutant.

From single-route optimization to AI-guided multifunctional design

A further priority is to move beyond single-route optimization toward AI-guided design of multifunctional or complementary-function biochars for mixed-contaminant systems. Rather than seeking a single optimal active site for a given field condition, future models should aim to identify combinations of reactive moieties, mineral phases, or blended amendment strategies that broaden functional coverage across cationic and oxyanionic metals. Progress in this direction will depend on standardized descriptors of surface chemistry, pore architecture, and mineral loading, as well as mechanistically informed training data and long-term validation under heterogeneous field conditions. Such frameworks could ultimately support multi-objective optimization of remediation efficiency, stability, cost, and environmental safety.

Long-term field validation and aging mechanisms

Most current knowledge is derived from short-term incubation studies. There is an urgent need for multi-season field trials to decipher the aging mechanisms of engineered biochars. Understanding how surface oxidation, pore clogging, and interactions with soil organic matter affect the long-term stability of immobilized metals is crucial for regulatory approval and commercialization.

Take-home Message

The traditional trial-and-error paradigm in biochar remediation is increasingly insufficient for addressing field-scale complexity. The next frontier involves a paradigm shift toward intelligent precision. Future research should harness AI not only to optimize immobilization efficiency but also to predict crop safety and screen for ecological toxicity via computational toxicology. By integrating the “soil-plant-human” continuum into data-driven frameworks, it may be possible to engineer biochars that are not only effective in theory but also safe, predictable, and compliant in reality.

REFERENCES CITED

Abdul Halim, N. S. a., Vijayanathan, J., Abdullah, R., Yaacob, J. S., Mazlan, M. A., Ahmad, R., Elham, P., and Kassim, A. S. (2024). “Influence of different pyrolysis temperature on the characteristics of forestry waste biochar for sodium adsorption,” Journal of Material Cycles and Waste Management 26(2), 894-907. https://doi.org/10.1007/s10163-023-01867-6

Ahmad, I., Shah, G. M., Shahbaz, H. M., Jamal, A., Rashid, M. I., Saeed, M. F., Farooq, A., Ali, Q., and Ronga, D. (2025). “Combined application of compost and biogas residues reduces soil and pore water metal concentration and their residual toxicity to wheat seedlings,” Journal of Soil Science and Plant Nutrition 25, 2479–2491. https://doi.org/10.1007/s42729-025-02279-1

Ahmed, A., and Aidi, H. (2025). “Biochar for heavy metal remediation: mechanisms, modifications, and environmental applications,” Environmental Science and Pollution Research 32(34), 20288-20320. https://doi.org/10.1007/s11356-025-36886-3

Alam, M. S., Bishop, B., Chen, N., Safari, S., Warter, V., Byrne, J. M., Warchola, T., Kappler, A., Konhauser, K. O., and Alessi, D. S. (2020). “Reusable magnetite nanoparticles-biochar composites for the efficient removal of chromate from water,” Scientific Reports 10(1), article 19007. https://doi.org/10.1038/s41598-020-75924-7

Alfei, S., and Pandoli, O. G. (2024). “Biochar-derived persistent free radicals: A plethora of environmental applications in a light and shadows scenario,” Toxics 12(4), article 245. https://doi.org/10.3390/toxics12040245

Alsamhary, K. (2023). “Vermi-cyanobacterial remediation of cadmium-contaminated soil with rice husk biochar: An eco-friendly approach,” Chemosphere 311, article 136931. https://doi.org/10.1016/j.chemosphere.2022.136931

Angon, P. B., Islam, M. S., Kc, S., Das, A., Anjum, N., Poudel, A., and Suchi, S. A. (2024). “Sources, effects and present perspectives of heavy metals contamination: Soil, plants and human food chain,” Heliyon 2024, article e28357. https://doi.org/10.1016/j.heliyon.2024.e28357

Aziz, K. H. H., and Kareem, R. (2023). “Recent advances in water remediation from toxic heavy metals using biochar as a green and efficient adsorbent: A review,” Case Studies in Chemical and Environmental Engineering 2023, article 100495. https://doi.org/10.1016/j.cscee.2023.100495

Balakumar, S., Mahesh, N., Kamaraj, M., and Aravind, J. (2025). “Harnessing artificial intelligence for sustainable environmental remediation. A review,” International Journal of Environmental Science and Technology 22(13), 13189-13206. https://doi.org/10.1007/s13762-025-06528-9

Barszcz, W., Łożyńska, M., and Molenda, J. (2024). “Impact of pyrolysis process conditions on the structure of biochar obtained from apple waste,” Scientific Reports 14(1), article 10501. https://doi.org/10.1038/s41598-024-61394-8

Bolan, S., Sharma, S., Mukherjee, S., Kumar, M., Rao, C. S., Nataraj, K. C., Singh, G., Vinu, A., Bhowmik, A., Sharma, H., El-Naggar, A., Chang, S. X., Hou, D., Rinklebe, J., Wang, H., Siddique, K. H. M., Abbott, L. K., Kirkham, M. B., and Bolan, N. (2024). “Biochar modulating soil biological health: A review,” Science of The Total Environment 914, article 169585. https://doi.org/10.1016/j.scitotenv.2023.169585

Chen, H., Zhang, J., Tang, L., Su, M., Tian, D., Zhang, L., Li, Z., and Hu, S. (2019). “Enhanced Pb immobilization via the combination of biochar and phosphate solubilizing bacteria,” Environment International 127, 395-401. https://doi.org/10.1016/j.envint.2019.03.068

Chen, W., Meng, J., Han, X., Lan, Y., and Zhang, W. (2019). “Past, present, and future of biochar,” Biochar 1(1), 75-87. https://doi.org/10.1007/s42773-019-00008-3

Chen, H., Yang, X., Liu, Y., Lin, X., Wang, J., Zhang, Z., Li, N., Li, Y., and Zhang, Y. (2021). “KOH modification effectively enhances the Cd and Pb adsorption performance of N-enriched biochar derived from waste chicken feathers,” Waste Management 130, 82-92. https://doi.org/10.1016/j.wasman.2021.05.015

Chen, W.-H., Hoang, A. T., Nižetić, S., Pandey, A., Cheng, C. K., Luque, R., Ong, H. C., Thomas, S., and Nguyen, X. P. (2022). “Biomass-derived biochar: From production to application in removing heavy metal-contaminated water,” Process Safety and Environmental Protection 160, 704-733. https://doi.org/10.1016/j.psep.2022.02.061

Chen, Z., Ding, Y., Jiang, X., Duan, H., Ruan, X., Li, Z., and Li, Y. (2022). “Combination of UNMIX, PMF model and Pb-Zn-Cu isotopic compositions for quantitative source apportionment of heavy metals in suburban agricultural soils,” Ecotoxicology and Environmental Safety 234, article 113369. https://doi.org/10.1016/j.ecoenv.2022.113369

Chen, D., Ye, X., Jiang, Y., Xiao, W., Zhang, Q., Zhao, S., Shao, S., Gao, N., Huang, M., and Hu, J. (2022). “Continuously applying compost for three years alleviated soil acidity and heavy metal bioavailability in a soil-asparagus lettuce system,” Frontiers in Plant Science 13, article 972789. https://doi.org/10.3389/fpls.2022.972789

Chen, X., Li, L., Li, X., Kang, J., Xiang, X., Shi, H., and Ren, X. (2023). “Effect of biochar on soil-water characteristics of soils: A pore-scale study,” Water 15(10), article 1909. https://www.mdpi.com/2073-4441/15/10/1909

Chen, M.-W., Chang, M.-S., Mao, Y., Hu, S., and Kung, C.-C. (2023). “Machine learning in the evaluation and prediction models of biochar application: A review,” Science Progress 106(1), article 00368504221148842. https://doi.org/10.1177/00368504221148842

Chen, Z., Kamchoom, V., Leung, A. K., Xue, J., and Chen, R. (2024). “Influence of biochar on the water permeability of compacted clay subjected to freezing–thawing cycles,” Acta Geophysica 72(3), 2071-2081. https://doi.org/10.1007/s11600-023-01141-1

Chen, J., Zhou, J., Zheng, W., Leng, S., Ai, Z., Zhang, W., Yang, Z., Yang, J., Xu, Z., and Cao, J. (2024). “A complete review on the oxygen-containing functional groups of biochar: Formation mechanisms, detection methods, engineering, and applications,” Science of The Total Environment 2024, article 174081. https://doi.org/10.1016/j.scitotenv.2024.174081

Cui, H., Dong, T., Hu, L., Xia, R., Zhou, J., and Zhou, J. (2022). “Adsorption and immobilization of soil lead by two phosphate-based biochars and phosphorus release risk assessment,” Science of The Total Environment 824, article 153957. https://doi.org/10.1016/j.scitotenv.2022.153957

Cui, Z., Wang, Y., Wang, N., Ma, F., and Yuan, Y. (2024). “Effects of ageing on surface properties of biochar and bioavailability of heavy metals in soil,” Agriculture 14(9), article 1631. Retrieved from https://www.mdpi.com/2077-0472/14/9/1631

Dai, J., Zhang, H., Chen, N., Wang, Y., Wang, J., Yang, Z., and Fang, G. (2025). “Nitrogen configurations driving efficient elimination of Cr(VI) by nitrogen-doped biochar/zero-valent iron composites,” Chemical Engineering Journal 515, article 163655. https://doi.org/10.1016/j.cej.2025.163655

de Lima Veloso, V., da Silva, F. B. V., dos Santos, N. M., and do Nascimento, C. W. A. (2022). “Phytoattenuation of Cd, Pb, and Zn in a slag-contaminated soil amended with rice straw biochar and grown with energy maize,” Environmental Management 69(1), 196-212. https://doi.org/10.1007/s00267-021-01530-6

Dhiman, A., Ramanthan, A., Macklin, M., Yadav, S., Kushwaha, S., Mudbhatkal, A., and Senapathi, V. (2023). “Heavy metal distribution in various environmental matrices and their risk assessment in Ganga River Basin, India,” Human and Ecological Risk Assessment: An International Journal 29(2), 621-650. https://doi.org/10.1080/10807039.2022.2159785

Dimitriadou, S., Isari, E. A., Grilla, E., Kokkinos, P., and Kalavrouziotis, I. K. (2025). “Revitalizing degraded soils: The role of biochar in enhancing soil health and productivity,” Environments 12(9), article 12090324. https://doi.org/10.3390/environments12090324

Dong, M., Jiang, M., He, L., Zhang, Z., Gustave, W., Vithanage, M., Niazi, N. K., Chen, B., Zhang, X., Wang, H., and He, F. (2025). “Challenges in safe environmental applications of biochar: Identifying risks and unintended consequence,” Biochar 7(1), article 12. https://doi.org/10.1007/s42773-024-00412-4

Dong, Q., Song, C., Yang, D., Zhao, Y., and Yan, M. (2023). “Spatial distribution, contamination assessment and origin of soil heavy metals in the Danjiangkou Reservoir, China,” International Journal of Environmental Research and Public Health 20(4), article 3443. https://doi.org/10.3390/ijerph20043443

El-Wakil, A. M., Waly, S. M., Abou El-Maaty, W. M., Waly, M. M., Yılmaz, M., and Awad, F. S. (2022). “Triazine-based functionalized activated carbon prepared from water hyacinth for the removal of Hg²⁺, Pb²⁺, and Cd²⁺ ions from water,” ACS Omega 7(7), 6058-6069. https://doi.org/10.1021/acsomega.1c06441

Elemile, O. O., Gana, A. J., Ejigboye, P. O., Ibitogbe, E. M., Olajide, O. S., and Ibitoye, O. O. (2023). “Analysis of potentially toxic elements from selected mechanical workshops using the geo-accumulation index and principal component analysis in Omu-Aran Community, Nigeria,” Environmental Monitoring and Assessment 195(2), article 276. https://doi.org/10.1007/s10661-022-10800-7

Fakhar, A., Galgo, S. J. C., Canatoy, R. C., Rafique, M., Sarfraz, R., Farooque, A. A., and Khan, M. I. (2025). “Advancing modified biochar for sustainable agriculture: A comprehensive review on characterization, analysis, and soil performance,” Biochar 7(1), article 8. https://doi.org/10.1007/s42773-024-00397-0

Faloye, O. T., Ajayi, E. A., Rostek, J., Schroeren, V., Babalola, T., Fashina, A., and Horn, R. (2024). “Hydraulic and pore functions of differently textured soils modified by biochar from different parts of the mango plant,” Soil and Tillage Research 236, article 105944. https://doi.org/10.1016/j.still.2023.105944

Fan, X., Wang, X., Zhao, B., Wan, J., Tang, J., and Guo, X. (2022). “Sorption mechanisms of diethyl phthalate by nutshell biochar derived at different pyrolysis temperature,” Journal of Environmental Chemical Engineering 10(2), article 107328. https://doi.org/10.1016/j.jece.2022.107328

Fei, X., Lou, Z., Xiao, R., Ren, Z., and Lv, X. (2022). “Source analysis and source-oriented risk assessment of heavy metal pollution in agricultural soils of different cultivated land qualities,” Journal of Cleaner Production 341, article 130942. https://doi.org/10.1016/j.jclepro.2022.130942

Flórez, E., Jimenez-Orozco, C., and Acelas, N. (2024). “Unravelling the influence of surface functional groups and surface charge on heavy metal adsorption onto carbonaceous materials: An in-depth DFT study,” Materials Today Communications 39, article 108647. https://doi.org/10.1016/j.mtcomm.2024.108647

Freiesleben, T., König, G., Molnar, C., and Tejero-Cantero, Á. (2024). “Scientific inference with interpretable machine learning: Analyzing models to learn about real-world phenomena,” Minds and Machines 34(3), article 32. https://doi.org/10.1007/s11023-024-09691-z

Garg, A., Chauhan, P., Kaur, C., Perveen, S., Arora, P. K., Garg, S. K., Singh, V. P., and Srivastava, A. (2025). “Bacterial bioremediation strategies for heavy metal detoxification: A multidisciplinary approach,” Environmental Sustainability 8, 395-417. https://doi.org/10.1007/s42398-025-00369-5

Ge, Y., Ying, K., Yu, G., Ali, M. U., Idris, A. M., Shahab, A., and Ullah, H. (2025). “A systematic review on machine learning-aided design of engineered biochar for soil and water contaminant removal,” Frontiers in Soil Science 5, article 1659154. https://doi.org/10.3389/fsoil.2025.1623083

Gong, H., Zhao, L., Rui, X., Hu, J., and Zhu, N. (2022). “A review of pristine and modified biochar immobilizing typical heavy metals in soil: Applications and challenges,” Journal of Hazardous Materials 432, article 128668. https://doi.org/10.1016/j.jhazmat.2022.128668

Grafmüller, J., Rathnayake, D., Hagemann, N., Bucheli, T. D., and Schmidt, H.-P. (2024). “Biochars from chlorine-rich feedstock are low in polychlorinated dioxins, furans and biphenyls,” Journal of Analytical and Applied Pyrolysis 183, article 106764. https://doi.org/10.1016/j.jaap.2024.106764

Guo, G., Lin, L., Jin, F., Mašek, O., and Huang, Q. (2023). “Application of heavy metal immobilization in soil by biochar using machine learning,” Environmental Research 231, article 116098. https://doi.org/10.1016/j.envres.2023.116098

Guo, G., Wang, Y., Zhang, D., Li, K., and Lei, M. (2023). “Human health risk apportionment from potential sources of heavy metals in agricultural soils and associated uncertainty analysis,” Environmental Geochemistry and Health 45(3), 881-897. https://doi.org/10.1007/s10653-022-01243-7

Guo, M., Song, W., and Tian, J. (2020). “Biochar-facilitated soil remediation: Mechanisms and efficacy variations,” Frontiers in Environmental Science 8, article 521512. https://doi.org/10.3389/fenvs.2020.521512

Guo, Z., Huang, L., Yan, J., Zhang, H., Jia, X., Li, M., and Li, H. (2025). “Artificial intelligence technology in environmental research and health: Development and prospects,” Environment International 203, article 109788. https://doi.org/10.1016/j.envint.2025.109788

Gusiatin, M. Z., and Rouhani, A. (2023). “Application of selected methods to modify pyrolyzed biochar for the immobilization of metals in soil: A review,” Materials 16(23), article 7342. https://www.mdpi.com/1996-1944/16/23/7342

Hafeez, A., Rasheed, R., Ashraf, M. A., Qureshi, F. F., Hussain, I., and Iqbal, M. (2023). “Effect of heavy metals on growth, physiological and biochemical responses of plants,” in: Plants and Their Interaction to Environmental Pollution, A. Husen (ed.), Elsevier, Ch. 8, pp. 139-159. https://doi.org/10.1016/B978-0-323-99978-6.00006-6

Hai, A., Bharath, G., Patah, M. F. A., Daud, W. M. A. W., K, R., Show, P., and Banat, F. (2023). “Machine learning models for the prediction of total yield and specific surface area of biochar derived from agricultural biomass by pyrolysis,” Environmental Technology & Innovation 30, article 103071. https://doi.org/10.1016/j.eti.2023.103071

Haider, F. U., Wang, X., Farooq, M., Hussain, S., Cheema, S. A., Ain, N. u., Virk, A. L., Ejaz, M., Janyshova, U., and Liqun, C. (2022). “Biochar application for the remediation of trace metals in contaminated soils: Implications for stress tolerance and crop production,” Ecotoxicology and Environmental Safety 230, article 113165. https://doi.org/10.1016/j.ecoenv.2022.113165

Hale, S. E., Lehmann, J., Rutherford, D., Zimmerman, A. R., Bachmann, R. T., Shitumbanuma, V., O’Toole, A., Sundqvist, K. L., Arp, H. P. H., and Cornelissen, G. (2012). “Quantifying the total and bioavailable polycyclic aromatic hydrocarbons and dioxins in biochars,” Environmental Science & Technology 46(5), 2830-2838. https://doi.org/10.1021/es203984k

Han, G.-F., Li, F., Zou, W., Karamad, M., Jeon, J.-P., Kim, S.-W., Kim, S.-J., Bu, Y., Fu, Z., Lu, Y., Siahrostami, S., and Baek, J.-B. (2020). “Building and identifying highly active oxygenated groups in carbon materials for oxygen reduction to H₂O₂,” Nature Communications 11(1), article 2209. https://doi.org/10.1038/s41467-020-15782-z

He, M., Xu, Z., Hou, D., Gao, B., Cao, X., Ok, Y. S., Rinklebe, J., Bolan, N. S., and Tsang, D. C. W. (2022). “Waste-derived biochar for water pollution control and sustainable development,” Nature Reviews Earth & Environment 3(7), 444-460. https://doi.org/10.1038/s43017-022-00306-8

He, X., Wang, Q., Jin, Y., Chen, Y., and Huang, L. (2024). “Properties of biochar colloids and behaviors in the soil environment: Influencing the migration of heavy metals,” Environmental Research 247, article 118340. https://doi.org/10.1016/j.envres.2024.118340

Hou, R., Zhu, B., Wang, L., Gao, S., Wang, R., and Hou, D. (2024). “Mechanism of clay mineral modified biochar simultaneously immobilizes heavy metals and reduces soil carbon emissions,” Journal of Environmental Management 361, article 121252. https://doi.org/10.1016/j.jenvman.2024.121252

Ibrahim, E. A., El-Sherbini, M. A. A., and Selim, E.-M. M. (2022). “Effects of biochar on soil properties, heavy metal availability and uptake, and growth of summer squash grown in metal-contaminated soil,” Scientia Horticulturae 301, article 111097. https://doi.org/10.1016/j.scienta.2022.111097

Ji, M. Y., Wang, X. X., Usman, M., Liu, F. H., Dan, Y. T., Zhou, L., Campanaro, S., Luo, G., and Sang, W. J. (2022). “Effects of different feedstocks-based biochar on soil remediation: A review,” Environmental Pollution 294(19), article 118655. https://doi.org/10.1016/j.envpol.2021.118655

Jia, M., Yu, J., Li, Z., Wu, L., and Christie, P. (2021). “Effects of biochar on the migration and transformation of metal species in a highly acid soil contaminated with multiple metals and leached with solutions of different pH,” Chemosphere 278, article 130344. https://doi.org/10.1016/j.chemosphere.2021.130344

Jiang, Y., Xie, S., Abou-Elwafa, S. F., Mukherjee, S., Singh, R. K., Tran, H.-T., Shi, J., Trindade, H., Zhang, T., and Chen, Q. (2025). “Machine learning-enabled optimization of biochar resource utilization and carbon mitigation pathways: Mechanisms and challenges,” Biochar X 1(1),  e002.  https://doi.org/10.48130/bchax-0025-0003

Jin, Y., Wang, Y., Li, X., Luo, T., Ma, Y., Wang, B., and Liang, H. (2023). “Remediation and its biological responses to Cd(II)-Cr(VI)-Pb(II) multi-contaminated soil by supported nano zero-valent iron composites,” Science of The Total Environment 867, article 161344. https://doi.org/10.1016/j.scitotenv.2022.161344

Joshi, M., Bhatt, D., and Srivastava, A. (2023). “Enhanced adsorption efficiency through biochar modification: A comprehensive review,” Industrial & Engineering Chemistry Research 62(35), 13748-13761. https://doi.org/10.1021/acs.iecr.3c02368

Joshi, N. C., and Gururani, P. (2023). “A mini review on heavy metal contamination in vegetable crops,” International Journal of Environmental Analytical Chemistry 2023, 1-12. https://doi.org/10.1080/03067319.2023.2210058

Kasera, N., Kolar, P., and Hall, S. G. (2022). “Nitrogen-doped biochars as adsorbents for mitigation of heavy metals and organics from water: A review,” Biochar 4(1), article 17. https://doi.org/10.1007/s42773-022-00145-2

Kicińska, A., Pomykała, R., and Izquierdo‐Diaz, M. (2022). “Changes in soil pH and mobility of heavy metals in contaminated soils,” European Journal of Soil Science 73(1), article e13203. https://doi.org/10.1111/ejss.13203

Krebsbach, S., He, J., Adhikari, S., Olshansky, Y., Feyzbar, F., Davis, L. C., Oh, T.-S., and Wang, D. (2023). “Mechanistic understanding of perfluorooctane sulfonate (PFOS) sorption by biochars,” Chemosphere 330, article 138661. https://doi.org/10.1016/j.chemosphere.2023.138661

Lai, W., Wu, Y., Zhang, C., Dilinuer, Y., Pasang, L., Lu, Y., Wang, Y., Chen, H., and Li, Z. (2022). “Combination of biochar and phosphorus solubilizing bacteria to improve the stable form of toxic metal minerals and microbial abundance in lead/cadmium-contaminated soil,” Agronomy 12(5), article 1003. https://www.mdpi.com/2073-4395/12/5/1003

Lawal, K. K., Ekeleme, I. K., Onuigbo, C. M., Ikpeazu, V. O., and Obiekezie, S. O. (2021). “A review on the public health implications of heavy metals,” World Journal of Advanced Research and Reviews 10(3), 255-265. https://doi.org/10.30574/wjarr.2021.10.3.0249

Leng, L., Xiong, Q., Yang, L., Li, H., Zhou, Y., Zhang, W., Jiang, S., Li, H., and Huang, H. (2021). “An overview on engineering the surface area and porosity of biochar,” Science of The Total Environment 763, article 144204. https://doi.org/10.1016/j.scitotenv.2020.144204

Li, H., Ai, Z., Yang, L., Zhang, W., Yang, Z., Peng, H., and Leng, L. (2023). “Machine learning assisted predicting and engineering specific surface area and total pore volume of biochar,” Bioresource Technology 369, article 128417. https://doi.org/10.1016/j.biortech.2022.128417

Li, H., Gao, P., Cui, J., Zhang, F., Wang, F., and Cheng, J. (2018). “Preparation and Cr(VI) removal performance of corncob activated carbon,” Environmental Science and Pollution Research 25(21), 20743-20755. https://doi.org/10.1007/s11356-018-2026-y

Li, J., Dai, J., Liu, G., Zhang, H., Gao, Z., Fu, J., He, Y., and Huang, Y. (2016). “Biochar from microwave pyrolysis of biomass: A review,” Biomass and Bioenergy 94, 228-244. https://doi.org/10.1016/j.biombioe.2016.09.010

Li, X., Chen, B., Chen, W., Yin, Y., Huang, L., Wei, L., Awad, M., and Liu, Z. (2024). “Predictive machine learning model to assess the adsorption efficiency of biochar-heavy metals for effective remediation of soil–plant environment,” Toxics 12(8), article 575. https://doi.org/10.3390/toxics12080575

Li, X., Liu, H., Meng, W., Liu, N., and Wu, P. (2022). “Accumulation and source apportionment of heavy metal(loid)s in agricultural soils based on GIS, SOM and PMF: A case study in superposition areas of geochemical anomalies and zinc smelting, Southwest China,” Process Safety and Environmental Protection 159, 964-977. https://doi.org/10.1016/j.psep.2022.01.072

Li, Y., Gupta, R., and You, S. (2022). “Machine learning assisted prediction of biochar yield and composition via pyrolysis of biomass,” Bioresource Technology 359, article 127511. https://doi.org/10.1016/j.biortech.2022.127511

Li, Y., and Gong, X. (2021). “Effects of dissolved organic matter on the bioavailability of heavy metals during microbial dissimilatory iron reduction: A review,” in: Reviews of Environmental Contamination and Toxicology, P. de Voogt (ed.), Springer International Publishing, Vol. 257, pp. 69-92. https://doi.org/10.1007/398_2020_63

Li, Z. L., Tang, Z., Song, Z. P., Chen, W. N., Tian, D. S., Tang, S. M., Wang, X. Y., Wang, J. S., Liu, W. J., Wang, Y., Li, J., Jiang, L. F., Luo, Y. Q., and Niu, S. L. (2022). “Variations and controlling factors of soil denitrification rate,” Global Change Biology 28(6), 2133-2145. https://doi.org/10.1111/gcb.16066

Li, Z., Huang, Y., Zhu, Z., Yu, M., Cheng, H., Shi, H., Xiao, Y., Song, H., Zuo, W., Zhou, H., and Wang, S. (2024). “Attempts to obtain clean biochar from hyperaccum-ulator through pyrolysis: Removal of heavy metals and transformation of phosphor-us,” Journal of Hazardous Materials 468, article 133837. https://doi.org/10.1016/j.jhazmat.2024.133837

Lian, W., Shi, W., Tian, S., Gong, X., Yu, Q., Lu, H., Liu, Z., Zheng, J., Wang, Y., Bian, R., Li, L., and Pan, G. (2023). “Preparation and application of biochar from co-pyrolysis of different feedstocks for immobilization of heavy metals in contaminated soil,” Waste Management 163, 12-21. https://doi.org/10.1016/j.wasman.2023.03.022

Lin, Y., Cai, Q., Chen, B., and Garg, A. (2025). “A review of the negative effects of biochar on soil in green infrastructure with consideration of soil properties,” Indian Geotechnical Journal 55(4), 2563-2577. https://doi.org/10.1007/s40098-024-00875-z

Lin, Y., Li, H., Li, X., and Wu, Y. (2024). “The stabilizing effect of chitosan-modified biochar (β-CD/C-BC) on agricultural soil contaminated by the Cd-Pyr complex,” ChemistrySelect 9(13), article e202304398. https://doi.org/10.1002/slct.202304398

Ling, L.-L., Liu, W.-J., Zhang, S., and Jiang, H. (2017). “Magnesium oxide embedded nitrogen self-doped biochar composites: Fast and high-efficiency adsorption of heavy metals in an aqueous solution,” Environmental Science & Technology 51(17), 10081-10089. https://doi.org/10.1021/acs.est.7b02382

Liu, L., Zhang, S., Chen, M., Fei, C., Zhang, W., Li, Y., and Ding, X. (2023). “Fe-modified biochar combined with mineral fertilization promotes soil organic phosphorus mineralization by shifting the diversity of phoD-harboring bacteria within soil aggregates in saline-alkaline paddy soil,” Journal of Soils and Sediments 23(2), 619-633. https://doi.org/10.1007/s11368-022-03335-4

Liu, M., Tao, J., Mu, L., Su, H., Peng, H., Cheng, Z., and Chen, G. (2025). “Machine learning-driven predictions of biochar yield and NPK composition: Insights into biomass pyrolysis with data augmentation and model interpretability,” Carbon Research 4(1), article 62. https://doi.org/10.1007/s44246-025-00229-1

Liu, R., Zhang, Y., Hu, B., and Wang, H. (2022). “Improved Pb(II) removal in aqueous solution by sulfide@biochar and polysaccharose-FeS@ biochar composites: Efficiencies and mechanisms,” Chemosphere 287, article 132087. https://doi.org/10.1016/j.chemosphere.2021.132087

Liu, S., Cen, B., Yu, Z., Qiu, R., Gao, T., and Long, X. (2025). “The key role of biochar in amending acidic soil: Reducing soil acidity and improving soil acid buffering capacity,” Biochar 7(1), article 52. https://doi.org/10.1007/s42773-025-00432-8

Liu, Y. H., Liu, G. Q., Zhang, J. L., Li, H. D., and Wu, J. (2023). “Effects of biosolid biochar on crop production and metal accumulation through a rice-wheat rotation system in fields,” Environmental Pollutants and Bioavailability 35(1), 9, article 2240016. https://doi.org/10.1080/26395940.2023.2240016

Liu, J., Kang, H., Tao, W., Li, H., He, D., Ma, L., Tang, H., Wu, S., Yang, K., and Li, X. (2023). “A spatial distribution – Principal component analysis (SD-PCA) model to assess pollution of heavy metals in soil,” Science of the Total Environment 859, article 160112. https://doi.org/10.1016/j.scitotenv.2022.160112

Long, X. X., Yu, Z. N., Liu, S. W., Gao, T., and Qiu, R. L. (2024). “A systematic review of biochar aging and the potential eco-environmental risk in heavy metal contaminated soil,” Journal of Hazardous Materials 472(22), article 134345. https://doi.org/10.1016/j.jhazmat.2024.134345

Lu, Y., Yin, Y., Shen, Z., Shi, B., and Gu, K. (2025). “Biochar aging: A review on its implications for soil engineering properties,” Bulletin of Engineering Geology and the Environment 84(11), article 561. https://doi.org/10.1007/s10064-025-04548-w

Lundberg, S. M., and Lee, S.-I. (2017). “A unified approach to interpreting model predictions,” Advances in Neural Information Processing Systems 30, 3295222-3295230. https://doi.org/10.5555/3295222.3295230

Luo, D., Wang, L., Nan, H., Cao, Y., Wang, H., Kumar, T. V., and Wang, C. (2023). “Phosphorus adsorption by functionalized biochar: A review,” Environmental Chemistry Letters 21(1), 497-524. https://doi.org/10.1007/s10311-022-01519-5

Luo, M., Jiang, X., Liu, Y., Liu, Y., Yu, H., Niu, Y., Meng, X., Wang, L., and Niu, Y. (2023). “Enhanced adsorption complexation of biochar by nitrogen-containing functional groups,” Journal of Environmental Chemical Engineering 11(6), article 111194. https://doi.org/10.1016/j.jece.2023.111194

Luo, Q. R., Liu, C. E., Zhao, Y. Q., Zhang, W. L., Chen, J. Q., Pan, Y., and Duan, C. Q. (2024). “Influence of biochar on the remediation of cadmium-contaminated soil in an Eisenia fetida–Solanum nigrum system,” Environmental Chemistry and Ecotoxicology 6, 132-143. https://doi.org/10.1016/j.enceco.2024.04.003

Luo, J., He, Y., Ning, W., Cao, M., and Hu, Y. (2023). “Unraveling the complexities of beryllium contamination in agricultural soils: The case of Qingcheng District, Qingyuan City,” Ecotoxicology and Environmental Safety 262, article 115142. https://doi.org/10.1016/j.ecoenv.2023.115142

Manikandan, S. K., Pallavi, P., Shetty, K., Bhattacharjee, D., Giannakoudakis, D. A., Katsoyiannis, I. A., and Nair, V. (2023). “Effective usage of biochar and microorganisms for the removal of heavy metal ions and pesticides,” Molecules 28(2), article 719. https://www.mdpi.com/1420-3049/28/2/719

Manning, J. R. H., Donval, G., Tolladay, M., Underwood, T. L., Parker, S. C., and Düren, T. (2023). “Identifying pathways to metal–organic framework collapse during solvent activation with molecular simulations,” Journal of Materials Chemistry A 11(47), 25929-25937. https://doi.org/10.1039/D3TA04647H

Micó, C., Recatalá, L., Peris, M., and Sánchez, J. (2006). “Assessing heavy metal sources in agricultural soils of an European Mediterranean area by multivariate analysis,” Chemosphere 65(5), 863-872. https://doi.org/10.1016/j.chemosphere.2006.03.016

Miftahushudur, T., Sahin, H. M., Grieve, B., and Yin, H. (2025). “A survey of methods for addressing imbalance data problems in agriculture applications,” Remote Sensing 17(3), article 454. Retrieved from https://www.mdpi.com/2072-4292/17/3/454

Mukherjee, S., Sarkar, B., Aralappanavar, V. K., Mukhopadhyay, R., Basak, B. B., Srivastava, P., Marchut-Mikołajczyk, O., Bhatnagar, A., Semple, K. T., and Bolan, N. (2022). “Biochar-microorganism interactions for organic pollutant remediation: Challenges and perspectives,” Environmental Pollution 308, article 119609. https://doi.org/10.1016/j.envpol.2022.119609

Murdoch, W. J., Singh, C., Kumbier, K., Abbasi-Asl, R., and Yu, B. (2019). “Definitions, methods, and applications in interpretable machine learning,” Proceedings of the National Academy of Sciences 116(44), 22071-22080. https://doi.org/doi:10.1073/pnas.1900654116

Murtaza, G., Ahmed, Z., and Usman, M. (2022). “Feedstock type, pyrolysis temperature and acid modification effects on physiochemical attributes of biochar and soil quality,” Arabian Journal of Geosciences 15(3), article 305. https://doi.org/10.1007/s12517-022-09539-9

Mwelwa, S., Chungu, D., Tailoka, F., Beesigamukama, D., and Tanga, C. (2023). “Biotransfer of heavy metals along the soil-plant-edible insect-human food chain in Africa,” Science of the Total Environment 881, article 163150. https://doi.org/10.1016/j.scitotenv.2025.178757

Nguyen, V. G., Sharma, P., Ağbulut, Ü., Le, H. S., Truong, T. H., Dzida, M., Tran, M. H., Le, H. C., and Tran, V. D. (2024). “Machine learning for the management of biochar yield and properties of biomass sources for sustainable energy,” Biofuels, Bioproducts and Biorefining 18(2), 567-593. https://doi.org/10.1002/bbb.2596

O’Connor, D., Peng, T., Zhang, J., Tsang, D. C., Alessi, D. S., Shen, Z., Bolan, N. S., and Hou, D. (2018). “Biochar application for the remediation of heavy metal polluted land: a review of in situ field trials,” Science of The Total Environment 619, 815-826. https://doi.org/10.1016/j.scitotenv.2017.11.132

Palansooriya, K. N., Li, J., Dissanayake, P. D., Suvarna, M., Li, L., Yuan, X., Sarkar, B., Tsang, D. C. W., Rinklebe, J., Wang, X., and Ok, Y. S. (2022). “Prediction of soil heavy metal immobilization by biochar using machine learning,” Environmental Science & Technology 56(7), 4187-4198. https://doi.org/10.1021/acs.est.1c08302

Peiris, C., Alahakoon, Y. A., Malaweera Arachchi, U., Mlsna, T. E., Gunatilake, S. R., and Zhang, X. (2023). “Phosphorus-enriched biochar for the remediation of heavy metal contaminated soil,” Journal of Agriculture and Food Research 12, article 100546. https://doi.org/10.1016/j.jafr.2023.100546

Qi, C., Xu, M., Liu, J., Li, C., Yang, B., Jin, Z., Liang, S., and Guo, B. (2024). “Source analysis and contribution estimation of heavy metal contamination in agricultural soils in an industrial town in the Yangtze river delta, China,” Minerals 14(3), article 279. https://doi.org/10.3390/min14030279

Qiao, P., Wang, S., Li, J., Zhao, Q., Wei, Y., Lei, M., Yang, J., and Zhang, Z. (2023). “Process, influencing factors, and simulation of the lateral transport of heavy metals in surface runoff in a mining area driven by rainfall: A review,” Science of the Total Environment 857, article 159119. https://doi.org/10.1016/j.scitotenv.2022.159119

Qin, G., Niu, Z., Yu, J., Li, Z., Ma, J., and Xiang, P. (2021). “Soil heavy metal pollution and food safety in China: Effects, sources and removing technology,” Chemosphere 267, article 129205. https://doi.org/10.1016/j.chemosphere.2020.129205

Rai, P. K., Sonne, C., and Kim, K.-H. (2023). “Heavy metals and arsenic stress in food crops: Elucidating antioxidative defence mechanisms in hyperaccumulators for food security, agricultural sustainability, and human health,” Science of the Total Environment 874, article 162327. https://doi.org/10.1016/j.scitotenv.2023.162327

Ramezanzadeh, H., Zarehaghi, D., Baybordi, A., Bouket, A. C., Oszako, T., Alenezi, F. N., and Belbahri, L. (2023). “The impacts of biochar-assisted factors on the hydrophysical characteristics of amended soils: A review,” Sustainability 15(11), 25, article 8700. https://doi.org/10.3390/su15118700

Ravindiran, G., Rajamanickam, S., Janardhan, G., Hayder, G., Alagumalai, A., Mahian, O., Lam, S. S., and Sonne, C. (2024). “Production and modifications of biochar to engineered materials and its application for environmental sustainability: A review,” Biochar 6(1), article 62. https://doi.org/10.1007/s42773-024-00350-1

Ren, J., Huang, H., Zhang, Z., Xu, X., Zhao, L., Qiu, H., and Cao, X. (2023). “Enhanced microbial reduction of Cr(VI) in soil with biochar acting as an electron shuttle: Crucial role of redox-active moieties,” Chemosphere 328, article 138601. https://doi.org/10.1016/j.chemosphere.2023.138601

Rehman, S. u., De Castro, F., Marini, P., Aprile, A., Benedetti, M., and Fanizzi, F. P. (2023). “Vermibiochar: A novel approach for reducing the environmental impact of heavy metals contamination in agricultural land,” Sustainability 15(12), article 9380. https://www.mdpi.com/2071-1050/15/12/9380

Rothwell, K. A., and Cooke, M. P. (2015). “A comparison of methods used to calculate normal background concentrations of potentially toxic elements for urban soil,” Science of The Total Environment 532, 625-634. https://doi.org/10.1016/j.scitotenv.2015.06.083

Saidi, P., Dasarathy, G., and Berisha, V. (2025). “Unraveling overoptimism and publication bias in ML-driven science,” Patterns 6(4), article 101185. https://doi.org/10.1016/j.patter.2025.101185

Schmidt, H.-P., Kammann, C., Hagemann, N., Leifeld, J., Bucheli, T. D., Sánchez Monedero, M. A., and Cayuela, M. L. (2021). “Biochar in agriculture – A systematic review of 26 global meta-analyses,” GCB Bioenergy 13(11), 1708-1730. https://doi.org/10.1111/gcbb.12889

Serrano, M. F., López, J. E., Henao, N., and Saldarriaga, J. F. (2024). “Phosphorus-loaded biochar-assisted phytoremediation to immobilize cadmium, chromium, and lead in soils,” ACS Omega 9(3), 3574-3587. https://doi.org/10.1021/acsomega.3c07433

Shan, B., Hao, R., Xu, H., Li, J., Li, Y., Xu, X., and Zhang, J. (2021). “A review on mechanism of biomineralization using microbial-induced precipitation for immobilizing lead ions,” Environmental Science and Pollution Research 28(24), 30486-30498. https://doi.org/10.1007/s11356-021-14045-8

Shi, T., Zhang, J., Shen, W., Wang, J., and Li, X. (2022). “Machine learning can identify the sources of heavy metals in agricultural soil: A case study in northern Guangdong Province, China,” Ecotoxicology and Environmental Safety 245, article 114107. https://doi.org/10.1016/j.ecoenv.2022.114107

Singh, R., Yadav, R., Pandey, V., Singh, A., Singh, M., Shanker, K., and Khare, P. (2024). “Effect of biochar on soil microbial community, dissipation and uptake of chlorpyrifos and atrazine,” Biochar 6. https://doi.org/10.1007/s42773-024-00306-5

Sinha, P., Datar, A., Jeong, C., Deng, X., Chung, Y. G., and Lin, L.-C. (2019). “Surface area determination of porous materials using the Brunauer–Emmett–Teller (BET) method: Limitations and improvements,” The Journal of Physical Chemistry C 123(33), 20195-20209. https://doi.org/10.1021/acs.jpcc.9b02116

Sobol, Ł., Dyjakon, A., and Soukup, K. (2023). “Dioxins and furans in biochars, hydrochars and torreficates produced by thermochemical conversion of biomass: A review,” Environmental Chemistry Letters 21(4), 2225-2249. https://doi.org/10.1007/s10311-023-01600-7

Song, P., Ma, W., Gao, X., Ai, S., Wang, J., and Liu, W. (2022). “Remediation mechanism of Cu, Zn, As, Cd, and Pb contaminated soil by biochar-supported nanoscale zero-valent iron and its impact on soil enzyme activity,” Journal of Cleaner Production 378, article 134510. https://doi.org/10.1016/j.jclepro.2022.134510

Song, X., Zhang, Y., Cao, N., Sun, D., Zhang, Z., Wang, Y., Wen, Y., Yang, Y., and Lyu, T. (2021). “Sustainable chromium (VI) removal from contaminated groundwater using nano-magnetite-modified biochar via rapid microwave synthesis,” Molecules (Basel, Switzerland) 26(1), article 103. https://www.mdpi.com/1420-3049/26/1/103

Souza, K. D., de Oliveira, M. A., Alcantara, G. U., Paulino, G. M., de Lima, R. P., Ferreira, O. E., Bezerra, A. D., Pimenta, L. P. S., and Machado, A. R. T. (2023). “Effect of pyrolysis temperature on the properties of the coffee grounds biochar and composition of its leachates,” Chemical Papers 77(7), 3947-3956. https://doi.org/10.1007/s11696-023-02755-x

Subramanian, P., Pandian, K., Pakkiyam, S., and others. (2025). “Biochar for heavy metal cleanup in soil and water: A review,” Biomass Conversion and Biorefinery 15, 11421-11441. https://doi.org/10.1007/s13399-024-05989-1

Sun, L., Gong, P., Sun, Y., Qin, Q., Song, K., Ye, J., Zhang, H., Zhou, B., and Xue, Y. (2022). “Modified chicken manure biochar enhanced the adsorption for Cd²⁺ in aqueous and immobilization of Cd in contaminated agricultural soil,” Science of the Total Environment 851, article 158252. https://doi.org/10.1016/j.scitotenv.2022.158252

Sun, P., Chen, Y., Liu, J., Lu, S., Guo, J., Zhang, Z., and Zheng, X. (2022). “Quantitative evaluation of the synergistic effect of biochar and plants on immobilization of Pb,” Journal of Environmental Management 316, article 115200. https://doi.org/10.1016/j.jenvman.2022.115200

Sun, T., Gao, G., Yang, W., Sun, Y., Huang, Q., Wang, L., and Liang, X. (2024). “High-efficiency remediation of Hg and Cd co-contaminated paddy soils by Fe–Mn oxide modified biochar and its microbial community responses,” Biochar 6(1), 57. https://doi.org/10.1007/s42773-024-00346-x

Sun, T., Yang, W., Xu, Y., Wang, L., Liang, X., Huang, Q., and Sun, Y. (2023). “Effect of Ca–modified biochar coupling with low–Cd accumulation maize cultivars on remediation of Cd contaminated soils and microbial community composition,” Soil and Tillage Research 232, article 105765. https://doi.org/10.1016/j.still.2023.105765

Sun, X., Wang, J., Zhang, M., Liu, Z., E, Y., Lan, Y., He, T., and Meng, J. (2023). “Effects of biochar on the Cd uptake by rice and the Cd fractions in paddy soil: A 3-year field experiment,” Agronomy 13(5), article 1335. https://www.mdpi.com/2073-4395/13/5/1335

Sun, Y., Jia, J., Huo, L., Zhao, L., Yao, Z., and Liu, Z. (2024). “Heteroatom-doped biochar for CO₂ adsorption: A review of heteroatoms, doping methods, and functions,” Biomass Conversion and Biorefinery 14(14), 15237-15249. https://doi.org/10.1007/s13399-022-03640-5

Šurda, P., Vitková, J., Lichner, Ľ., Botková, N., and Toková, L. (2024). “Effect of wettable and hydrophobic biochar addition on properties of sandy soil,” Biologia 80, 1247-1258. https://doi.org/10.1007/s11756-024-01702-9

Tan, W.-T., Zhou, H., Tang, S.-F., Chen, Q., Zhou, X., Liu, X.-H., Zeng, P., Gu, J.-F., and Liao, B.-H. (2023). “Simultaneous alleviation of Cd availability in contaminated soil and accumulation in rice (Oryza sativa L.) by Fe-Mn oxide-modified biochar,” Science of The Total Environment 858, article 159730. https://doi.org/10.1016/j.scitotenv.2022.159730

Tao, Q., Li, B., Chen, Y. X., Zhao, J. W., Li, Q. Q., Chen, Y. L., Peng, Q. H., Yuan, S., Li, H. X., Huang, R., and Wang, C. Q. (2021). “An integrated method to produce fermented liquid feed and biologically modified biochar as cadmium adsorbents using corn stalks,” Waste Management 127, 112-120. https://doi.org/10.1016/j.wasman.2021.04.027

Uppalapati, S., Paramasivam, P., Kilari, N., Chohan, J. S., Kanti, P. K., Vemanaboina, H., Dabelo, L. H., and Gupta, R. (2025). “Precision biochar yield forecasting employing random forest and XGBoost with Taylor diagram visualization,” Scientific Reports 15(1), article 7105. https://doi.org/10.1038/s41598-025-91450-w

Vabalas, A., Gowen, E., Poliakoff, E., and Casson, A. J. (2019). “Machine learning algorithm validation with a limited sample size,” PLOS One 14(11), article e0224365. https://doi.org/10.1371/journal.pone.0224365

Venkatachalam, C. D., Sekar, S., Sengottian, M., Ravichandran, S. R., and Bhuvaneshwaran, P. (2023). “A critical review of the production, activation, and morphological characteristic study on functionalized biochar,” Journal of Energy Storage 67, article 107525. https://doi.org/10.1016/j.est.2023.107525

Vijay, V., Shreedhar, S., Adlak, K., Payyanad, S., Sreedharan, V., Gopi, G., van der Voort, S. T., Malarvizhi, P., Yi, S., Gebert, J., and Aravind, P. (2021). “Review of large-scale biochar field-trials for soil amendment and the observed influences on crop yield variations,” Frontiers in Energy Research 9, article 710166. https://doi.org/10.3389/fenrg.2021.710766

Wang, A. D., Zou, D. S., O’Connor, P., Chen, B., Zou, J. C., Zhou, Y. Q., Wang, H., and Zhang, M. Y. (2023). “Assistant effects of spent mushroom substrate and its derived biochar on soil phytoremediation,” Journal of Soils and Sediments 23(4), 1641-1653. https://doi.org/10.1007/s11368-023-03452-8

Wang, J., Norgaard, T., Pugliese, L., Carvalho, P. N., and Wu, S. (2024). “Global meta-analysis and machine learning reveal the critical role of soil properties in influencing biochar-pesticide interactions,” Environment International 193, article 109131. https://doi.org/10.1016/j.envint.2024.109131

Wang, J., Shi, L., Liu, J., Deng, J., Zou, J., Zhang, X., Shen, Z., and Chen, Y. (2023). “Earthworm-mediated nitrification and gut digestive processes facilitate the remobilization of biochar-immobilized heavy metals,” Environmental Pollution 322, article 121219. https://doi.org/10.1016/j.envpol.2023.121219

Wang, L., Wang, Y., Ma, F., Tankpa, V., Bai, S., Guo, X., and Wang, X. (2019). “Mechanisms and reutilization of modified biochar used for removal of heavy metals from wastewater: A review,” Science of The Total Environment 668, 1298-1309. https://doi.org/10.1016/j.scitotenv.2019.03.011

Wang, Q., Xu, C., Pan, K., Wu, X., Pan, Y., Duan, C., and Geng, Z. (2025). “P-modified biochar alters the microbial community in heavy metal-contaminated soils by regulating nutrient supply balance,” Biochar 7(1), article 93. https://doi.org/10.1007/s42773-025-00495-7

Wang, S., Li, J., Yuan, X., Senadheera, S. S., Chang, S. X., Wang, X., and Ok, Y. S. (2024). “Machine learning predicts biochar aging effects on nitrous oxide emissions from agricultural soils,” ACS Agricultural Science & Technology 4(9), 888-898. https://doi.org/10.1021/acsagscitech.4c00114

Wang, Y. C., Ni, J. J., Guo, H. W., and Kravchenko, E. (2024). “Influences of phosphorus-modified biochar on bacterial community and diversity in rhizosphere soil,” Environmental Science and Pollution Research 31(1), 1681-1691. https://doi.org/10.1007/s11356-023-31556-8

Weidemann, E., Buss, W., Edo, M., Mašek, O., and Jansson, S. (2018). “Influence of pyrolysis temperature and production unit on formation of selected PAHs, oxy-PAHs, N-PACs, PCDDs, and PCDFs in biochar—A screening study,” Environmental Science and Pollution Research 25(4), 3933-3940. https://doi.org/10.1007/s11356-017-0612-z

Wen, C., Liu, T., Wang, D., Wang, Y., Chen, H., Luo, G., Zhou, Z., Li, C., and Xu, M. (2023). “Biochar as the effective adsorbent to combustion gaseous pollutants: Preparation, activation, functionalization and the adsorption mechanisms,” Progress in Energy and Combustion Science 99, article 101098.

Wu, D., Feng, Z., Gu, W., Wang, Y., Liu, Z., Wang, W., Zhang, Y., Zhang, W., and Chen, W. (2023). “Could continuous rice cropping increase soil fertility and rice productivity by rice straw carbonized utilization in cold areas? — A 6-year field-located trial,” Environmental Science and Pollution Research 30(51), 110674-110686. https://doi.org/10.1007/s11356-023-30097-4

Xiang, L., Harindintwali, J. D., Wang, F., Redmile-Gordon, M., Chang, S. X., Fu, Y., He, C., Muhoza, B., Brahushi, F., Bolan, N., Jiang, X., Ok, Y. S., Rinklebe, J., Schaeffer, A., Zhu, Y.-g., Tiedje, J. M., and Xing, B. (2022). “Integrating biochar, bacteria, and plants for sustainable remediation of soils contaminated with organic pollutants,” Environmental Science & Technology 56(23), 16546-16566. https://doi.org/10.1021/acs.est.2c02976

Xin, Z., Zhao, B., Liu, H., Zhao, Y., and Li, L. (2023). “Mechanisms of sludge biochar effects on thermal properties of a loess soil (Sierozem),” Eurasian Soil Science 56(11), 1683-1695. https://doi.org/10.1134/S1064229323600689

Xing, J., Dong, W., Liang, N., Huang, Y., Wu, M., Zhang, L., and Chen, Q. (2023). “Sorption of organic contaminants by biochars with multiple porous structures: Experiments and molecular dynamics simulations mediated by three-dimensional models,” Journal of Hazardous Materials 458, article 131953. https://doi.org/10.1016/j.jhazmat.2023.131953

Xu, D., Shen, Z., Dou, C., Dou, Z., Li, Y., Gao, Y., and Sun, Q. (2022). “Effects of soil properties on heavy metal bioavailability and accumulation in crop grains under different farmland use patterns,” Scientific Reports 12(1), article 9211. https://doi.org/10.1038/s41598-022-13140-1

Xu, J., Li, Y., and Li, L. (2025). “A comprehensive review of the effects of organic amendments on soil health and fertility: Mechanisms, greenhouse gas emissions, and implications for sustainable agriculture,” Agronomy, 15(12), article 2705. Retrieved from https://www.mdpi.com/2073-4395/15/12/2705

Xu, Q., and Du, L. (2024). “Using machine learning to predict Biochar yield and carbon content: enhancing efficiency and sustainability in biomass conversion,” BioResources 19(3), 6545-6558. https://doi.org/10.15376/biores.19.3.6545-6558

Xu, Z., Xu, X., Yu, Y., Yao, C., Tsang, D. C. W., and Cao, X. (2021). “Evolution of redox activity of biochar during interaction with soil minerals: Effect on the electron donating and mediating capacities for Cr(VI) reduction,” Journal of Hazardous Materials 414, article 125483. https://doi.org/10.1016/j.jhazmat.2021.125483

Yakout, S. M., Daifullah, A. E. H. M., and El-Reefy, S. A. (2015). “Pore structure characterization of chemically modified biochar derived from rice straw,” Environmental Engineering & Management Journal (EEMJ) 14(2), article 49. https://doi.org/10.30638/EEMJ.2015.049

Yang, M., Luo, S., Zeng, P., and Wu, Y. (2025). “Fabrications and properties of heteroatom-based co-doped biochar for environmental application: A review,” Separations 12(2), article 20. https://doi.org/10.3390/separations12020020

Yang, X. Y., Zhang, Z. M., Zhang, J. C., Wu, X. L., Luo, W. M., and Mu, G. T. (2023). “Effect of biochar and Acidithiobacillus ferrooxidans on heavy metal content in soil and lettuce,” Soil & Sediment Contamination 32(7), 910-925. https://doi.org/10.1080/15320383.2022.2154746

Yao, C., Wang, B., Zhang, J., Faheem, M., Feng, Q., Hassan, M., Zhang, X., Lee, X., and Wang, S. (2024). “Formation mechanisms and degradation methods of polycyclic aromatic hydrocarbons in biochar: A review,” Journal of Environmental Management 357, article 120610. https://doi.org/10.1016/j.jenvman.2024.120610

Yin, M., Bai, X., Wu, D., Li, F., Jiang, K., Ma, N., Chen, Z., Zhang, X., and Fang, L. (2022). “Sulfur-functional group tunning on biochar through sodium thiosulfate modified molten salt process for efficient heavy metal adsorption,” Chemical Engineering Journal 433, article 134441. https://doi.org/10.1016/j.cej.2021.134441

Yuan, Z., Wang, Y., Zhu, L., Zhang, C., and Sun, Y. (2024). “Machine-learning-aided biochar production from aquatic biomass,” Carbon Research 3(1), article 77. https://doi.org/10.1007/s44246-024-00169-2

Yu, Q., Wen, J., Zhang, S., Wu, C., Ouyang, H., Hu, N., Li, X., and Qiu, X. (2024). “The coupling of sulfide and Fe-Mn mineral promotes the migration of lead and zinc in the redox cycle of high pH floodplain soils,” Journal of Hazardous Materials 472, article 134546. https://doi.org/10.1016/j.jhazmat.2024.134546

Zand, A. D., and Abyaneh, M. R. (2020). “Adsorption of cadmium from landfill leachate on wood-derived biochar: Non-linear regression analysis,” Environmental Processes 7(4), 1129-1150. https://doi.org/10.1007/s40710-020-00461-4

Zhang, C., Liu, H., Zhou, F., Long, X., Liu, S., and Wu, Y. (2024). “Enhancing remediation efficiency of hyperaccumulators through earthworm addition: Evidence from a pot study on cadmium-contaminated soil,” Science of the Total Environment 934, article 173169. https://doi.org/10.1016/j.scitotenv.2024.173169

Zhang, H., Zhang, Y., Bai, L., Zhang, Y., and Sun, L. (2021). “Effect of physiochemical properties in biomass-derived materials caused by different synthesis methods and their electrochemical properties in supercapacitors,” Journal of Materials Chemistry A 9(21), 12521-12552. https://doi.org/10.1039/D1TA00790D

Zhang, M. R., Liu, Z., Han, F., Cong, H. T., and Zhou, W. Z. (2024). “Co-application of phytoremediation with iron-loaded biochar in petroleum and zinc co-contaminated soil,” Environmental Research 263, 12, article 120037. https://doi.org/10.1016/j.envres.2024.120037

Zhang, P., Li, Y., Qi, H., and Li, X. (2023). “Adsorption characteristics of water ciprofloxacin on co-pyrolysis biochar derived from rubber and ginkgo biloba leaves,” ChemistrySelect 8(34), article e202302116. https://doi.org/10.1002/slct.202302116

Zhang, Q., and Wang, C. (2020). “Natural and human factors affect the distribution of soil heavy metal pollution: A review,” Water, Air, & Soil Pollution 231(7), article 350. https://doi.org/10.1007/s11270-020-04728-2

Zhang, Y., Ren, M., Tang, Y., Cui, X., Cui, J., Xu, C., Qie, H., Tan, X., Liu, D., Zhao, J., Wang, S., and Lin, A. (2022). “Immobilization on anionic metal(loid)s in soil by biochar: A meta-analysis assisted by machine learning,” Journal of Hazardous Materials 438, article 129442. https://doi.org/10.1016/j.jhazmat.2022.129442

Zhao, J., Guo, R., Guo, J., Yu, Z., Xu, J., Zhang, X., and Yang, L. (2025). “Spatiotemporal patterns and driving mechanisms of heavy metal accumulation in China’s farmland soils based on meta-analysis and machine learning,” Sustainability 17(24), article 11318. Retrieved from https://www.mdpi.com/2071-1050/17/24/11318

Zhao, Z., Zhu, S., Qi, S., Zhou, T., Yang, Y., Wang, F., Han, Q., Dong, W., Wang, H., and Sun, F. (2025). “Collaborative modification strategy to improve the formation of biochar-derived persistent free radicals for aniline removal via peroxymonosulfate activation,” Biochar 7(1), article 24. https://doi.org/10.1007/s42773-024-00416-0

Zheng, Q., Yang, L. F., Song, D. L., Zhang, S., Wu, H., Li, S. T., and Wang, X. B. (2020). “High adsorption capacity of Mg-Al-modified biochar for phosphate and its potential for phosphate interception in soil,” Chemosphere 259, article 127469. https://doi.org/10.1016/j.chemosphere.2020.127469

Zhou, X., Liu, X., Sun, L., Jia, X., Tian, F., Liu, Y., and Wu, Z. (2024). “Prediction of biochar yield and specific surface area based on integrated learning algorithm. C 10(1), article 10. https://doi.org/10.3390/c10010010

Zhu, X., Li, Y., and Wang, X. (2019). “Machine learning prediction of biochar yield and carbon contents in biochar based on biomass characteristics and pyrolysis conditions,” Bioresource Technology 288, article 121527. https://doi.org/10.1016/j.biortech.2019.121527

Zhu, Q., Ji, J., Tang, X., Wang, C., and Sun, H. (2023). “Bioavailability assessment of heavy metals and organic pollutants in water and soil using DGT: A review,” Applied Sciences 13(17), article 9760. https://www.mdpi.com/2076-3417/13/17/9760

Article submitted: January 15, 2026; Peer review completed: March 21, 2026; Revised version received: April 10, 2026; Accepted: April 12, 2026; Published: May 26, 2026.

DOI: 10.15376/biores.21.3.Wu

SUPPLEMENTARY INFORMATION

Table S1. Origins of Heavy Metals in Agricultural Soil

Table S2. Physicochemical Properties of Biochar Prepared from Different Feedstocks Under Various Pyrolysis Conditions

Table S3. Recent Studies on Agricultural Soil Remediation by Biochar

Table S4. Summary of AI/ML Applications in Biochar Research