Effects of rice husk biochar and its modified forms on Cd²⁺ immobilization, humus dynamics, and soil fertility in Cd-contaminated paddy soils

Summary

With the widespread Cd²⁺ contamination of paddy soils threatening agricultural sustainability and food safety, this study explored the remediation potential of raw rice husk biochar (BC) and chemically modified BCs (BC-HCl, BC-NaOH, and BC-KMnO₄) for Cd-contaminated paddy soils. A 120-day incubation experiment was conducted across four Cd²⁺ gradients (0, 10, 20, and 40 mg/kg), and analytical techniques were used to characterize BC microstructure, Cd²⁺ immobilization efficiency, humus (HS) dynamics, and soil fertility parameters. Chemical modifications improved BC properties: BC-KMnO₄ had the highest –COOH content and degree of aromatization, while BC-NaOH showed strong alkalinity. At Cd²⁺ concentrations ≤20 mg/kg, BC-KMnO₄ performed best in Cd²⁺ immobilization and HS enhancement, whereas BC-NaOH was superior in stabilizing soil properties under severe Cd²⁺ contamination (40 mg/kg). This work clarified the synergistic mechanisms of BC in Cd passivation and soil fertility improvement and provided theoretical support for safe rice production in Cd-contaminated agricultural systems.

Graphical abstract

Highlights

• •BC-KMnO₄ had the highest –COOH and aromatization, BC-NaOH showed strong alkalinity
• •BC-KMnO₄ was optimal for Cd²⁺ immobilization and HS enhancement at Cd²⁺ ≤20 mg/kg
• •BC-NaOH stabilized soil properties well under 40 mg/kg Cd²⁺ stress
• •Modified BC promoted HS fraction accumulation and optimized HA aromatization degree

Soil science; Environmental science; Environmental management; Soil chemistry; Soil ecotoxicology

Introduction

With the acceleration of industrialization and urbanization in China, anthropogenic activities—notably mining, non-ferrous metal smelting, chemical manufacturing, and improper application of cadmium (Cd)-containing fertilizers and pesticides—have become dominant sources of Cd²⁺ pollution, leading to widespread and cumulative Cd²⁺ contamination in agricultural soils, especially severe in major rice-producing regions (e.g., the Songnen plain). For instance, Ke et al.¹ reported that an analysis of 484 rice samples collected from contaminated regions showed the average Cd²⁺ content across different areas ranged from 0.149 to 0.189 mg/kg, with over 18.0% of the samples exceeding the maximum allowable Cd²⁺ concentration for rice. Cd is notable for its persistent non-biodegradability, high potential to bioaccumulate, and pronounced chronic toxicity. These characteristics not only deteriorate soil structure, inhibit soil microbial processes, and reduce rice yield and grain quality but also heighten health risks for humans via food chain bioaccumulation, potentially inducing skeletal lesions and renal dysfunction.² Against this backdrop, there is an urgent need to develop remediation technologies that are efficient, cost-effective, and environmentally sustainable, which has become a critical priority in ecological and environmental protection. Biochar (BC), a carbon-rich, highly porous material produced via biomass pyrolysis, is widely recognized for heavy metal adsorption due to its diverse and abundant feedstocks, low production cost, rich surface functional groups, and long-term environmental persistence.³ Nevertheless, unmodified BC is limited by its relatively small specific surface area (SSA), insufficient number of active sites, and insufficient selectivity toward particular heavy metals, ultimately limiting its remediation performance. To address the above limitations, researchers have developed chemical modification methods to enhance the efficacy of BC in remediating Cd-contaminated soils.³

Recent studies have increasingly confirmed the pivotal role of BC modification in remediating heavy metal-contaminated environments. Li et al.⁴ demonstrated that the hydrothermal carbonization of BC increased its aromaticity and enriched surface functional groups, while anaerobic pyrolysis yielded BCs with enhanced porosity and larger surface areas. Their work further showed that combining magnetic BC with ryegrass reduced soil concentrations of Cr, Ni, Cu, Zn, As, and Cd by 24.12%, 23.30%, 22.01%, 9.98%, 14.83%, and 15.08%, respectively, along with corresponding reductions in their bioavailable fractions. Liu et al.⁵ evaluated iron-manganese-modified BC and confirmed its efficacy in removing heavy metals, organic matter (OM), phosphates, and nitrates through mechanisms including mesoporous adsorption, redox reactions, complexation, electrostatic attraction, and precipitation. In a related pot experiment, Sun et al.⁶ found that sulfur-modified chicken manure BC significantly increased soil pH, reduced soil Cd²⁺ bioavailability, and decreased Cd²⁺ uptake in Brassica chinensis shoots, indicating its potential for safe utilization of Cd-contaminated farmland. Tan et al.⁷ reported that acid-pickling BC exhibited a marked reduction in Cd²⁺ adsorption capacity, likely due to the removal of mineral phases and a subsequent decrease in soil pH. Further, Ali et al.⁸ showed that apricot shell and apple tree-derived BCs facilitated the transformation of Cd²⁺ and Zn²⁺ from acid-soluble and reducible fractions to organic-bound and residual fractions, which typically pose less bioavailability. Rosales et al.⁹ identified that the removal mechanism of Cd²⁺ by metal-modified BC involved cation-π bonding, ion exchange, and surface complexation, while Mosa et al.¹⁰ highlighted that metal modification might attenuate the electrostatic adsorption of Cd²⁺. Additionally, Ahuekwe et al.¹¹ associated the post-KOH modification increase in MgO, K₂O, and Al₂O₃ contents with enhanced heavy metal immobilization efficiency.

Despite these advancements, most existing studies have concentrated on a single or a limited range of modifications. There remains a lack of systematic comparisons regarding the remediation efficiency and soil fertility preservation effects of various modified BCs under consistent experimental conditions, particularly in Cd-contaminated agricultural soils. To address this research gap, this study systematically evaluated four representative BC types (BC, BC-HCl, BC-NaOH, and BC-KMnO₄) across four Cd²⁺ contamination gradients to assess their performance in mitigating Cd²⁺ pollution while maintaining soil function. A multi-method analytical framework was employed, integrating scanning electron microscopy (SEM) for characterizing BC microstructural features, atomic absorption spectrophotometry for quantifying Cd²⁺ accumulation in different rice organs, and Fourier transform infrared spectroscopy (FTIR) for elucidating the functional group composition of humic acid (HA). This study additionally employed principal component analysis (PCA) for dimensionality reduction of evaluation metrics, thereby facilitating a holistic assessment of experimental performance. The optimal amendment was screened based on its combined Cd²⁺ remediation and soil fertility improvement effects under 0–40 mg/kg Cd²⁺ contamination. Additionally, this work revealed the synergistic mechanisms of Cd²⁺ immobilization, soil humification, and fertility maintenance regulated by BC microstructures and surface functional groups, providing scientific support for the safe utilization of Cd-contaminated paddy soils.

Results

SEM micrographs and FTIR spectra of BC and its chemically modified BCs

To elucidate the microstructural characteristics of BC and chemically modified BC samples, SEM analysis was performed for morphological characterization (Figure 1), while the SSA, pore volume, and average pore size were measured (Table 1). As shown in the SEM micrographs, BC exhibited a relatively smooth surface morphology, accompanied by a discontinuous macroporous structure and incomplete carbonization. The SSA, pore volume, and average pore size of BC were determined to be 72.8 m²/g, 0.086 cm³/g, and 3.15 nm, respectively. BC-HCl not only cleansed the BC surface but also exposed the underlying C matrix, resulting in a significant increase in SSA to 142.2 m²/g. Correspondingly, its pore volume and pore size increased to 0.097 cm³/g and 3.31 nm, respectively. The SSA, pore volume, and average pore size of BC-NaOH were measured as 127.3 m²/g, 0.107 cm³/g, and 4.71 nm, respectively. Nevertheless, the SSA of BC-KMnO₄ was still significantly enhanced to 128.8 m²/g, while its pore volume and pore size were 0.112 cm³/g and 3.03 nm, respectively.

Figure 1.

SEM micrographs of BC and its chemically modified BCs

Note: The scale bar represented 50 μm, and the image was captured at a magnification of 1,000×.

Table 1.

Specific surface area, pore volume, and pore size of BC and chemically modified BCs

Types of BC	SSA (m2/g)	Pore volume (cm3/g)	Pore size (nm)
BC (Ⅰ)	72.8 c	0.086 days	3.15 c
BC-HCl (Ⅱ)	142.2 a	0.097 c	3.31 b
BC-NaOH (Ⅲ)	127.3 b	0.107 b	4.71 a
BC-KMnO4 (Ⅳ)	128.8 b	0.112 a	3.03 days

Lowercase letters denote statistically significant differences among BC types at p < 0.05, as determined by one-way analysis of variance (ANOVA) followed by post-hoc least significant difference (LSD) and Tukey’s honestly significant difference (HSD, S-B(k)) tests. The sample size for each group was n = 3. The analytical method employed in this differential analysis was consistently applied to all subsequent differential analyses presented in this argument. Henceforth, the same method was adopted throughout.

FTIR spectroscopy was employed to further characterize the structural features and surface functional group composition of BC and its chemically modified BCs (BC-HCl, BC-NaOH, BC-KMnO₄), based on the positions and relative intensities of key absorption peaks.¹² Key FTIR spectral features were summarized in Table 2 and Figure 2. A broad absorption peak in the range of 3425–3434 cm⁻¹ was attributed to the stretching vibration of hydroxyl groups (–OH).¹³ After chemical modification, the intensity of the –OH band followed the order: BC-NaOH (44.0) > BC-KMnO₄ (43.1) > BC-HCl (42.2) > BC (41.5). The absorption band at 1738 cm⁻¹, corresponding to carboxyl group (–COOH), exhibited a comparable trend in –COOH content: BC-KMnO₄ (5.6) > BC-HCl (5.1) > BC-NaOH (4.9) > BC (4.7).¹⁴ To quantify the aromatization degree of BC, the (a+b)/c ratio, representing the relative content of aliphatic C to aromatic C, was used.¹⁵ All chemical modifications led to a decrease in the (a+b)/c ratio relative to BC (0.74), with the respective values being 0.59 (BC-HCl), 0.52 (BC-NaOH), and 0.45 (BC-KMnO₄). Additionally, the absorption band in the range of 1084–1093 cm⁻¹ was assigned to the stretching vibration of O–Si–O bonds.¹⁶ In comparison to BC (17.0), the O-Si-O bond strength increased in BC-KMnO₄ (22.7) and BC-HCl (20.8), while BC-NaOH (12.3) displayed the most pronounced reduction in O–Si–O intensity.

Table 2.

Absorption peak intensities (expressed as a percentage of total spectral area) of BC and chemically modified BCs

Types of BC	3425–3434	2922–2924a	2852–2854b	1738	1595–1622c	1084–1093	(a+b)/c
Ⅰ BC	41.5 c	7.7 a	3.0 b	4.7 days	14.5 days	17.0 c	0.74 a
Ⅱ BC-HCl	42.2 b	7.3 b	2.6 days	5.1 b	19.2 a	20.8 b	0.59 b
Ⅲ BC-NaOH	44.0 a	7.8 a	2.8 c	4.9 c	18.1 b	12.3 days	0.52 c
Ⅳ BC-KMnO4	43.1 c	3.6 c	3.2 a	5.6 a	15.1 c	22.7 a	0.45 days

Figure 2.

FTIR spectra of BC and chemically modified BCs

Cd²⁺ concentration in different organs of rice plants

Figures 3A–3D shown that Cd²⁺ translocation from root to aerial parts (stem, leaf) and subsequent accumulation in grains were key to regulating Cd²⁺ levels in edible rice tissues.¹⁷ Across all tested rice plant organs, the order of Cd²⁺ accumulation consistently followed: BC > BC-HCl > BC-NaOH > BC-KMnO₄. Under exogenous Cd²⁺ stress conditions, Cd²⁺ concentrations in all rice tissues increased progressively with rising exogenous soil Cd²⁺ levels. Under exogenous Cd²⁺ concentrations of 10 mg/kg and 20 mg/kg (Cd10 and Cd20), the hierarchy of Cd²⁺ accumulation across all plant organs remained consistent with that under Cd0 (BC > BC-HCl > BC-NaOH > BC-KMnO₄). However, when the exogenous Cd²⁺ concentration was increased to 40 mg/kg (Cd40), this order shifted to BC > BC-HCl > BC-KMnO₄ > BC-NaOH.

Figure 3.

Cd²⁺ contents in different rice plant organs under four Cd²⁺ concentrations across BC and chemically modified BCs

Note: In the treatment groups, the values 0, 10, 20, and 40 stood for the added Cd²⁺ concentrations (mg/kg) shown in Figures (a), (b), (c), and (d), respectively; whereas Ⅰ, Ⅱ, Ⅲ, and Ⅳ corresponded to BC, BC-HCl, BC-NaOH, and BC-KMnO₄, respectively.

The same below. Data are represented as mean ± SEM.

Different uppercase letters indicate significant differences between BC types under the same Cd²⁺ concentration and within the same plant organ (p < 0.05).

Lowercase letters indicate significant differences across different organs under identical BC treatment and Cd²⁺ concentration. All statistical analyses were performed using one-way ANOVA with Duncan’s multiple range test as the post-hoc analysis; data were presented as mean ± SD, with n = 3 for each group. The analytical method employed in this difference analysis was uniformly adopted for all subsequent difference analyses in this argument.

Hereinafter, the same method was applied.

Soil pH value

Figure 4 summarized the variations in soil pH at 0 and 120 days of incubation, as influenced by exogenous Cd²⁺ concentrations and BC modifications. The regulation of soil pH by BC was critical for maintaining soil quality levels, and BC modification type exerted significant impacts on this regulatory capacity.¹⁸ In the absence of exogenous Cd²⁺, all chemically modified BC treatments significantly elevated soil pH relative to the BC. Specifically, BC-HCl exhibited the lowest alkalizing capacity, increasing pH by only 0.15; in contrast, BC-NaOH induced the greatest pH increase (0.88). Under exogenous Cd²⁺ stress, the pH-elevating efficacy of BCs declined progressively with increasing Cd²⁺ concentrations. Under exogenous 10 mg/kg Cd²⁺ stress, BC, BC-NaOH, and BC-KMnO₄ still increased soil pH by 0.17, 0.76, and 0.22 units, respectively; in contrast, BC-HCl slightly reduced pH by 0.09 units. Under exogenous 20 mg/kg Cd²⁺ stress, the pH increments further declined to 0.05, 0.40, and 0.01 units for BC, BC-NaOH, and BC-KMnO₄, respectively, while BC-HCl caused a more substantial pH decrease of 0.12 units. Under exogenous 40 mg/kg Cd²⁺ stress, only BC-NaOH maintained a modest alkalizing effect, with a 0.20-unit pH increase; all other BC modifications (BC, BC-HCl, and BC-KMnO₄) led to significant reductions in soil pH

Figure 4.

Soil pH values under four Cd²⁺ concentrations in BC and chemically modified BCs

Soil organic matter (OM) content

As illustrated in Figures 5A–5D, OM serves as a key indicator for assessing soil fertility and plays a critical role in influencing the physical, chemical, and biological properties of soil.¹⁹ In the absence of exogenous Cd²⁺ stress, OM content across all four BC modifications followed a consistent “rise-then-fall” trajectory, peaking at 40 days of incubation. By 40 days, BC-KMnO₄ exhibited the highest OM content. By 120 days, OM content remained above initial levels, with respective increments of 4.76% (BC), 5.81% (BC-HCl), 29.4% (BC-NaOH), and 25.6% (BC-KMnO₄). Notably, BC-NaOH and BC-KMnO₄ treatments demonstrated the strongest capacity for long-term OM sequestration. Under exogenous Cd²⁺ stress, OM content similarly followed a “rise-then-fall” pattern. Compared with the 0 mg/kg, the accumulation increase of OM was greater under the 10 mg/kg exogenous Cd²⁺ stress. Under exogenous concentrations of Cd²⁺ (10 and 20 mg/kg), all BC treatments significantly promoted the accumulation of OM, with the magnitude of increase following the order: BC-KMnO₄ > BC-NaOH > BC-HCl > BC. Under exogenous 40 mg/kg Cd²⁺ stress, the OM content in the BC-NaOH and BC-KMnO₄ treatments increased significantly by 18.3% and 12.5%; however, the OM content in the BC and BC-HCl treatments showed no statistically significant difference (p > 0.05).

Figure 5.

Effects of BC and chemically modified BCs on soil OM content under different Cd²⁺ concentrations

Effects of BC and chemically modified BCs on soil humus composition

As illustrated in Figures 6A–6D, the C content of HS was monitored at 0 and 120 days of incubation. Initially, the C content of water-soluble substances (C_WSS) remained low across all treatments, ranging from 0.19 to 0.30 g/kg. Under Cd²⁺ concentrations ≤20 mg/kg, all BC treatments significantly promoted C_WSS accumulation. The efficacy of BCs in promoting C_WSS accumulation followed the order: BC-KMnO₄ > BC-NaOH > BC-HCl > BC. Under exogenous 40 mg/kg Cd²⁺ stress, C_WSS levels significantly decreased under the BC, BC-HCl, and BC-KMnO₄ treatments by 23.1%, 16.0%, and 8.70%, respectively. In contrast, the BC-NaOH treatment continued to enhance C_WSS accumulation, with an increase of 42.9%; at 0 days, the C content of HA (C_HA) ranged from 0.91 to 1.04 g/kg. Under Cd²⁺ concentrations 0, 10, and 20 mg/kg, the C_HA accumulation rates followed the consistent rank order: BC-KMnO₄ > BC-NaOH > BC-HCl > BC. Under exogenous 40 mg/kg Cd²⁺ stress, C_HA decreased in most BC treatments; however, BC-NaOH still exhibited a small but positive change (8.05%). Initial C content in humic-extracted acid (C_HE) ranged from 3.14 to 3.33 g/kg. After 120 days, in the absence of exogenous Cd²⁺ addition, C_HE decreased in all BC treatments, with percentage reductions of 29.3% (BC-KMnO₄), 21.7% (BC-NaOH), 21.0% (BC-HCl), and 16.6% (BC). Under low exogenous concentrations of Cd²⁺ (10 mg/kg), BC showed the smallest reduction in C_HE, whereas BC-HCl, BC-NaOH, and BC-KMnO₄ exhibited more pronounced C_HE losses. Under 20 mg/kg exogenous concentrations of Cd²⁺, C_HE slightly increased in the BC and BC-HCl but continued to decline in BC-NaOH and BC-KMnO₄. Under exogenous 40 mg/kg Cd²⁺ stress, C_HE increased in BC and BC-HCl, stabilized in BC-KMnO₄, and further decreased in BC-NaOH. The initial C content of humin (C_Hu) ranged from 4.42 to 4.59 g/kg. After 120 days under exogenous 0 and 10 mg/kg Cd²⁺ stress, all BC treatments increased C_Hu, with BC-KMnO₄ showing the highest relative increase, followed by BC-NaOH, BC-HCl, and BC. Under 20 mg/kg exogenous concentrations of Cd²⁺, C_Hu decreased in all treatments: BC-KMnO₄ exhibited the smallest loss (0.45%), while BC-NaOH, BC-HCl, and BC showed larger reductions. Under exogenous 40 mg/kg Cd²⁺ stress, the loss of C_Hu was more pronounced in BC and BC-HCl, whereas BC-KMnO₄ maintained a minimal loss, and BC-NaOH in particular only showed a 0.22% reduction in C_Hu.

Figure 6.

Responses and variations of C_WSS, C_HE, C_HA, and C_Hu in soils amended with BC and chemically modified BCs under various Cd²⁺ concentrations

Note: Data are represented as mean ± SEM.

ΔlogK of HA and C_HA/C_FA ratio in soils amended with BC and chemically modified BCs under different Cd²⁺ concentrations

As illustrated in Figure 7A, the regulatory effects of Cd²⁺ concentration gradients and four BC types on the ΔlogK of HA were systematically investigated. A decreased ΔlogK indicated an increase in molecular weight and structural complexity of HA.²⁰ The ratio of C_HA/C_FA was an indicator of humification maturity, where higher C_HA/C_FA ratios signify enhanced HA enrichment and greater HS stability.²¹ Both ΔlogK and C_HA/C_FA showed distinct concentration-dependent responses to BC modification and Cd²⁺ stress. Under Cd²⁺ concentrations ≤20 mg/kg, the magnitude of ΔlogK reduction followed the order: BC-KMnO₄ > BC-NaOH > BC-HCl > BC. Notably, under 40 mg/kg exogenous concentrations of Cd²⁺, ΔlogK increased by 12.4% and 5.58% in the BC and BC-HCl treatments, respectively. In contrast, BC-NaOH and BC-KMnO₄ still induced decreases in ΔlogK of 27.4% and 19.5%, respectively.

Figure 7.

Effects of BC and chemically modified BCs on HS quality at varying Cd²⁺ concentrations

Note: Data are represented as mean ± SEM

As shown in Figure 7B, the C_HA/C_FA ratio was similarly regulated by BC type and Cd²⁺ concentration. The initial C_HA/C_FA ratio (0 days) ranged from 0.39 to 0.47 across all treatments. At Cd²⁺ concentrations ≤20 mg/kg, all BC amendments significantly increased the C_HA/C_FA ratio, with the highest increment observed in BC-KMnO₄, followed by BC-NaOH, BC-HCl, and BC. Under exogenous 10 and 20 mg/kg Cd²⁺ concentrations, C_HA/C_FA increments were attenuated relative to Cd0, but BC-KMnO₄ still retained the highest HA enrichment. At 40 mg/kg exogenous concentrations of Cd²⁺, the C_HA/C_FA ratio decreased markedly by 17.5% and 16.4% in the BC and BC-HCl treatments, respectively. In contrast, BC-KMnO₄ and BC-NaOH still enhanced the C_HA/C_FA ratio by 6.24% and 20.0%, respectively.

FTIR spectra and atomic molar ratios of HA

Comparative FTIR spectral analyses of HA were performed for all BC treatments at the 0 days and 120 days time points, as illustrated in Figures 8A–8H and Table 3. After 120 days, the intensity of functional groups and key structural ratios of HA demonstrated distinct, concentration-dependent responses to BC modification and exogenous Cd²⁺ concentration. In the absence of exogenous Cd²⁺ addition, all BC treatments significantly enhanced the intensity of –OH groups in HA, with the enhancement efficiency following the order: BC-KMnO₄ (45.6%) > BC-NaOH (34.6%) > BC-HCl (22.1%) > BC (10.9%). Two critical structural ratios of HA also displayed uniform decreasing patterns: the (a+b+c)/e ratio (reflecting the proportion of aliphatic C/carboxyl C and the structural complexity and condensation degree of HA) decreased by 7.48% (BC), 21.3% (BC-HCl), 51.7% (BC-NaOH), and 63.6% (BC-KMnO₄). Meanwhile, the (a+b+c)/d ratio (indicating the proportion of aliphatic C/aromatic C and the aromatization degree of HA) dropped by 20.9% (BC), 24.1% (BC-HCl), 44.5% (BC-NaOH), and 44.9% (BC-KMnO₄).²² At low exogenous concentrations of Cd²⁺ (10 mg/kg), all BC treatments increased –OH group intensity of HA, with enhancement levels following the order: BC-KMnO₄ (66.2%) > BC-NaOH (51.5%) > BC-HCl (36.9%) > BC (14.1%). Simultaneously, the (a+b+c)/e ratio decreased across all treatments, with reductions of 43.2% (BC), 35.8% (BC-HCl), 53.3% (BC-NaOH), and 55.8% (BC-KMnO₄); the (a+b+c)/d ratio also declined, showing reduction rates of 25.1% (BC), 41.9% (BC-HCl), 65.9% (BC-NaOH), and 64.8% (BC-KMnO₄). At 20 mg/kg exogenous concentrations of Cd²⁺, the enhancement of HA’s –OH intensity was significantly attenuated: BC even caused a 0.35% decrease in –OH intensity, whereas BC-KMnO₄ maintained the highest increase (53.5%), followed by BC-NaOH (36.1%) and BC-HCl (20.5%). The (a+b+c)/e ratio decreased by 14.1% (BC), 24.7% (BC-HCl), 47.9% (BC-NaOH), and 52.7% (BC-KMnO₄). The (a+b+c)/d ratio declined by 23.7% (BC), 35.5% (BC-HCl), 62.0% (BC-NaOH), and 63.6% (BC-KMnO₄). At the highest Cd²⁺ concentration (40 mg/kg), divergent responses emerged among the BC treatments. The –OH intensity of HA decreased by 21.7% (BC) and 10.2% (BC-HCl), but increased by 19.9% (BC-NaOH) and 1.78% (BC-KMnO₄). For the (a+b+c)/e ratio: BC exhibited a 4.07% increase, whereas BC-HCl, BC-NaOH, and BC-KMnO₄ showed decreases of 3.21%, 20.2%, and 5.66%, respectively. The (a+b+c)/d ratio increased by 42.0% in BC but decreased by 27.5% (BC-HCl), 57.3% (BC-NaOH), and 53.0% (BC-KMnO₄).

Figure 8.

FTIR spectra of HA in soils amended with BC and chemically modified BCs under varying Cd²⁺ concentrations

Table 3.

Relative FTIR intensities of HA (expressed as percentage of total spectral area) in soils amended with BC and chemically modified BCs at different Cd²⁺ concentrations

Rice growth period (d)	Treatments	3414–3434	2925–2983a	2852–2855b	1775–1778e	1600-1620 days	1432–1449c	(a+b+c)/d	(a+b+c)/e
0	Cd0-Ⅰ	28.4	2.8	1.4	4.4	8.0	16.4	2.58	4.68
120	Cd0-Ⅰ	31.5	2.9	3.9	3.9	9.2	10.1	2.05	4.33
0	Cd0-Ⅱ	28.1	3.6	3.5	4.2	7.7	12.4	2.53	4.64
120	Cd0-Ⅱ	34.3	3.4	4.1	6.5	11.1	13.8	1.92	3.65
0	Cd0-Ⅲ	28.3	3.1	3.9	3.1	5.5	6.50	2.45	4.35
120	Cd0-Ⅲ	38.1	3.8	3.2	7.2	8.4	4.40	1.36	2.10
0	Cd0-Ⅳ	29.4	4.9	4.3	4.7	7.7	11.2	2.65	4.34
120	Cd0-Ⅳ	42.8	2.1	1.3	4.1	5.9	5.20	1.46	1.58
Rice growth period (d)	Treatments	3141–3428	2922–2933a	2849–2852b	1775–1779e	1621–1678days	1441–1444c	(a+b+c)/d	(a+b+c)/e
0	Cd10-Ⅰ	28.4	3.7	3.7	4.3	8.3	11.8	2.31	4.47
120	Cd10-Ⅰ	32.4	2.4	1.8	7.2	10.6	14.1	1.73	2.54
0	Cd10-Ⅱ	27.1	5.2	3.0	4.6	8.4	12.6	2.48	4.52
120	Cd10-Ⅱ	37.1	4.2	3.8	6.1	12.3	9.70	1.44	2.90
0	Cd10-Ⅲ	29.1	3.2	3.6	4.5	7.8	13.8	2.64	4.58
120	Cd10-Ⅲ	44.1	2.7	2.5	5.1	12.1	5.70	0.90	2.14
0	Cd10-Ⅳ	27.8	4.7	5.4	4.8	8.5	11.4	2.53	4.48
120	Cd10-Ⅳ	46.2	3.4	0.4	6.3	14.1	8.70	0.89	1.98
Rice growth period (d)	Treatments	3414–3425	2925–2930a	2852–2855b	1775–1779e	1598–1675days	1441–1445c	(a+b+c)/d	(a+b+c)/e
0	Cd20-Ⅰ	28.2	3.3	5.4	4.4	7.8	11.0	2.53	4.48
120	Cd20-Ⅰ	28.1	3.1	5.1	4.1	9.7	7.6	1.93	3.85
0	Cd20-Ⅱ	28.3	4.5	3.9	3.8	6.7	8.4	2.51	4.42
120	Cd20-Ⅱ	34.1	3.9	4.2	5.2	14.2	9.2	1.62	3.33
0	Cd20-Ⅲ	28.5	2.9	1.2	4.9	9.2	18.2	2.42	4.55
120	Cd20-Ⅲ	38.8	1.1	2.6	4.1	10.5	6.0	0.92	2.37
0	Cd20-Ⅳ	28.2	4.2	3.6	4.8	9.1	13.7	2.36	4.48
120	Cd20-Ⅳ	43.3	1.5	2.5	4.9	12.1	6.4	0.86	2.12
Rice growth period (d)	Treatments	3414–3419	2922–2931a	2852b	1775–1778e	1601–1662days	1438–1452c	(a+b+c)/d	(a+b+c)/e
0	Cd40-Ⅰ	28.6	4.2	3.9	3.8	6.9	8.7	2.43	4.42
120	Cd40-Ⅰ	22.4	2.1	3.2	3.0	4.0	8.5	3.45	4.60
0	Cd40-Ⅱ	28.5	4.8	2.1	3.9	7.7	11.3	2.36	4.67
120	Cd40-Ⅱ	25.6	1.5	1.6	2.5	6.6	8.2	1.71	4.52
0	Cd40-Ⅲ	29.1	2.1	5.8	4.60	8.8	13.3	2.41	4.61
120	Cd40-Ⅲ	34.9	4.7	4.4	3.10	11.1	2.3	1.03	3.68
0	Cd40-Ⅳ	28.1	4.7	4.8	4.80	8.5	11.7	2.49	4.42
120	Cd40-Ⅳ	28.6	1.3	1.6	1.80	6.4	4.6	1.17	4.17

Table 4 presented the dynamic changes in the elemental atomic ratios (H/C, C/N, and O/C) of HA between the initial (0 days) and final (120 days) incubation stages, as influenced by different BC types and exogenous Cd²⁺ concentrations. The C/N ratio is a critical parameter for evaluating HA maturity and stability.²³ The H/C ratio is positively correlated with aliphatic group content and negatively correlated with aromatization degree.²⁴ The O/C ratio reflects the abundance of O-containing functional groups (e.g., –COOH, –OH, and C–O–C) in HA. In the absence of exogenous Cd²⁺ addition, all BC treatments led to a decrease in the H/C ratio, with reduction magnitudes following the order: BC-KMnO₄ (23.0%) > BC-NaOH (16.4%) > BC-HCl (13.1%) > BC (1.67%). Regarding the C/N ratio, unmodified BC showed a marginal increase of 0.59%, BC-HCl and BC-NaOH exhibited a 5.1% increase, and BC-KMnO₄ showed the highest increment at 17.7%. The O/C ratio increased significantly across all treatments, by 74.2% (BC), 75.8% (BC-HCl), 93.9% (BC-NaOH), and 102.8% (BC-KMnO₄). Under 10 mg/kg and 20 mg/kg exogenous Cd²⁺ stress, the trends in elemental molar ratios across treatments and the rank order of their relative increases remained consistent with those observed in the Cd²⁺-free treatment. Notably, under exogenous 40 mg/kg Cd²⁺ stress, contrasting responses in the H/C ratio were observed: BC and BC-HCl showed an increase of 39.3% and 20.3%, respectively, whereas BC-NaOH and BC-KMnO₄ decreased by 18.6% and 15.0%. All four BC treatments induced modest increases in the C/N ratio, ranging from 1.28% (BC) to 7.54% (BC-NaOH). The O/C ratio increased from 17.6% (BC) to 51.5% (BC-KMnO₄).

Table 4.

Atomic molar of HA in paddy soil amended with BC and chemically modified BCs under varying Cd²⁺ concentrations

Treatments	H/C ratio		C/N ratio		O/C ratio
–	0 days	120 days	0 days	120 days	0 days	120 days
Cd0-Ⅰ	0.60 b	0.59 a	16.84 b	16.94 c	0.31 c	0.54 days
Cd0-Ⅱ	0.61 a	0.53 b	16.82 b	17.67 b	0.33 b	0.58 c
Cd0-Ⅲ	0.61 a	0.51 c	16.75 c	17.60 b	0.33 b	0.64 b
Cd0-Ⅳ	0.61 a	0.47 days	17.04 a	20.05 a	0.36 a	0.73 a
Cd10-Ⅰ	0.60 b	0.57 a	16.50 days	17.19 days	0.29 days	0.56 days
Cd10-Ⅱ	0.60 b	0.54 b	17.00 a	17.83 c	0.31 c	0.60 c
Cd10-Ⅲ	0.61 a	0.54 b	16.75 c	18.10 b	0.32 b	0.74 b
Cd10-Ⅳ	0.60 b	0.43 c	16.92 b	20.40 a	0.37 a	0.82 a
Cd20-Ⅰ	0.60 a	0.56 a	16.73 b	17.22 days	0.34 a	0.46 c
Cd20-Ⅱ	0.60 a	0.55 b	16.35 days	17.46 c	0.29 c	0.49 c
Cd20-Ⅲ	0.60 a	0.53 c	16.51 c	17.64 b	0.31 b	0.58 b
Cd20-Ⅳ	0.60 a	0.49 days	16.92 a	19.43 a	0.31 b	0.69 a
Cd40-Ⅰ	0.61 a	0.85 a	16.38 days	16.59 c	0.34 b	0.40 days
Cd40-Ⅱ	0.59 c	0.71 b	16.65 b	16.91 b	0.36 a	0.45 c
Cd40-Ⅲ	0.59 c	0.48 days	16.44 c	17.68 a	0.34 b	0.47 b
Cd40-Ⅳ	0.60 b	0.51 c	16.88 a	16.94 b	0.33 c	0.50 a

Principal component analysis (PCA)

To comprehensively evaluate the effects of modified BCs on soil properties and Cd²⁺ accumulation in rice, PCA was performed using 20 indicators. At all Cd²⁺ concentrations, the first two principal components (PC1 and PC2) explained over 90% of the total variance (Table S1), indicating that PC1 and PC2 effectively captured the variation induced by different BC treatments. In the PCA without exogenous Cd²⁺ addition, PC1 (74.43%) and PC2 (16.01%) collectively accounted for 90.44% of the variance (Figures 9A; Table S1). As shown in Table S2, PC1 was primarily influenced by OM content, pore volume, C_WSS, and Cd²⁺ concentrations in rice stem, grain, and root—with all rice Cd²⁺ parameters showing negative loadings, while soil-related parameters showed positive loadings. PC2 was defined by soil pH, SSA, and the (a+b+c)/d ratio. The ranking of BC treatments based on PCA scores was BC-KMnO₄ > BC-NaOH > BC-HCl > BC (Table S3), with BC-KMnO₄ achieving the highest score (2.99). At 10 mg/kg exogenous concentrations of Cd²⁺, PC1 (79.88%) and PC2 (12.92%) explained 92.8% of the total variance (Figures 9B; Table S1). PC1 was most strongly associated with C_WSS, C_HA, pore volume, OM content, and grain Cd²⁺ concentration. PC2 mainly differentiated treatments based on pore size and soil pH (Table S2). BC-KMnO₄ again ranked first (score: 3.97), followed by BC-NaOH, BC-HCl, and raw BC (Table S3); At 20 mg/kg exogenous concentrations of Cd²⁺, PC1 (77.85%) and PC2 (14.71%) together accounted for 92.56% of the total variance (Figures 9C; Table S1). PC1 was driven by C_HA, pore volume, C_WSS, and Cd²⁺ concentrations in rice grains and stems, while PC2 was determined by soil pH and pore size (Table S2). The treatment ranking remained BC-KMnO₄ > BC-NaOH > BC-HCl > BC, with BC-KMnO₄ obtaining a score of 3.54 (Table S3); At 40 mg/kg exogenous concentrations of Cd²⁺, PC1 (73.32%) and PC2 (20.34%) jointly explained 93.67% of the total variance (Figures 9D; Table S1). PC1 was dominated by C_HA, C_HE, the C_HA/C_FA ratio, and Cd²⁺ concentrations in rice stem and root. PC2 differentiated treatments based on the (a+b+c)/d ratio, pore size, and soil pH (Table S2). Notably, BC-NaOH outperformed all other treatments at this Cd²⁺ level (score: 2.68), followed by BC-KMnO₄, BC-HCl, and BC (Table S3).

Figure 9.

Effects of BC and chemically modified BCs on Cd²⁺ immobilization and soil fertility: an integrated PCA-based insight

Discussion

Effects of unmodified and chemically modified BCs on their physicochemical properties and microstructural characteristics

Chemical modification of BCs substantially enhanced their microstructural, surface, and functional properties, thereby expanding their applicability for Cd²⁺ immobilization in paddy soils relative to BC. Specifically, BC-HCl treatment improved surface cleanliness and SSA by selectively dissolving acid-soluble minerals (e.g., carbonates, metal oxides, and silicates) and removing inorganic and organic surface impurities—thereby exposing the underlying C framework while preserving skeletal integrity and enhancing the accessibility of active sites.^25,26 BC-NaOH treatment enlarged the pore network, notably converting mesopores into macropores via alkaline etching; this yielded an open, honeycomb-like architecture with thinner pore walls and sharper pore edges, collectively increasing SSA.^27,28 Additionally, alkaline dissolution removed lignin-derived organic compounds and restructured the C skeleton, inducing the structural reorganization of the carbon skeleton and unveiling latent O-containing functional groups.²⁸ For BC-KMnO₄, a dense MnO_x layer was in situ deposited on the pore surfaces, which reduced pore size, but markedly increased redox-active sites, thereby promoting Cd²⁺ immobilization through coupled redox reactions and chemisorption.^27,29,30 Concurrently, the oxidative nature of KMnO₄ markedly increased the abundance of O-containing functional groups on BC, further augmenting sorption capacity for target pollutants such as Cd²⁺.³⁰

FTIR analysis confirmed that chemical modification increased O-containing functional groups and removed surface lipids.³¹ BC-NaOH exhibited the highest level of –OH group, activation by removing lignin derivatives to expose masked phenolic and alcoholic –OH group.²⁷ BC-KMnO₄ had the highest –COOH group, resulting from the oxidative conversion of surface –OH groups by KMnO₄; BC-HCl exposed pre-existing –COOH groups by dissolving mineral encrustations.^32,33 The polar groups (–OH, –COOH) facilitated Cd²⁺ immobilization through hydrogen bonding, ion exchange, and inner-sphere complexation.³⁴ Chemical modification also improved the aromatization degree of BC, a key indicator of C stability. BC-KMnO₄ exhibited the highest aromaticity due to aliphatic side chain cleavage and aromatic ring condensation.³⁵ BC-HCl enhanced aromatization by removing aliphatic impurities that diluted the aromatic C fraction, whereas BC-NaOH exposed aromatic domains via aliphatic lignin derivative dissolution.^36,37 This enhanced aromaticity improved BC chemical stability and Cd²⁺ affinity via π–π interactions.³⁸ Modification altered O–Si–O bond abundance, improving surface hydrophilicity and chemical heterogeneity to synergistically enhance interfacial reactivity toward Cd²⁺.³⁹ BC-KMnO₄ induced SiO₂ migration and precipitation, forming additional O–Si–O bonds.⁴⁰ BC-HCl exposed buried O–Si–O bonds by dissolving encapsulating minerals.⁴¹ BC-NaOH caused partial O–Si–O framework degradation, resulting in a smaller net increase in bond abundance.⁴²

Effects of unmodified and chemically modified BCs on regulating Cd²⁺ accumulation in different rice organs

The modified BCs further regulated Cd²⁺ accumulation in rice organs, and, regardless of Cd²⁺ stress presence, rice organs consistently exhibited a Cd²⁺ accumulation hierarchy: root > stem > leaf > grain, highlighting roots as a biological barrier that restricted Cd²⁺ translocation to aerial edible tissues and ensured grain safety.⁴³ In the absence of exogenous Cd²⁺, BC-KMnO₄ demonstrated the strongest efficacy in inhibiting root Cd²⁺ uptake, primarily due to surface in-situ-formed MnO_x groups that immobilized rhizospheric Cd²⁺ into iron-manganese oxide-bound fractions inaccessible to plants.^30,44,45 Additionally, oxidative modification by KMnO₄ increased the negative surface charge density of BC, which enhanced electrostatic attraction toward Cd²⁺ and consequently reduced the proportion of bioavailable Cd²⁺ in the soil.⁴⁶ BC-NaOH also had notable inhibitory effects, as alkaline etching exposed –OH and –COO^- groups for ligand exchange and inner-sphere complex formation with Cd²⁺, and its alkalinity elevated soil pH to promote Cd precipitation as CdCO₃.^34,47 BC-HCl was the least effective, as acid treatment dissolved minerals and reduced basic functional groups, decreasing BC surface negative charge and Cd²⁺ adsorption capacity.⁴⁸ At low-to-moderate exogenous Cd²⁺ concentrations (10–20 mg/kg), the hierarchy of Cd²⁺ accumulation was maintained across all BC treatments, indicating modified BC functional groups retained sufficient sorption capacity. Notably, when exogenous Cd²⁺ concentration reaches the highest level (40 mg/kg), BC-NaOH exhibited the strongest inhibitory effect. This shift was attributed to MnO_x binding site saturation under high Cd²⁺ loads, weakening BC-KMnO₄’s immobilization capacity, while BC-NaOH’s alkaline functional groups maintained robust sorption and complexation with Cd²⁺ even under severe stress.^28,49

Effects of unmodified and chemically modified BCs on soil pH and OM content under Cd²⁺ stress

Soil pH strongly governed Cd²⁺ adsorption onto BC by regulating BC surface charge density, surface functional group ionization, and Cd²⁺ speciation in soil solution.⁵⁰ Under neutral or alkaline conditions, deprotonated surface sites favored Cd²⁺ binding, while acidic conditions promoted protonation and raised electrostatic repulsion, inhibiting Cd²⁺ sorption.^51,52 In the absence of exogenous Cd²⁺, all BC treatments elevated soil pH, attributed to the inherent alkalinity of BC—derived from abundant surface –OH and –COOH groups, as well as soluble alkali metal cations (e.g., Na⁺ and K⁺).¹⁸ BC-NaOH had the strongest pH-elevating effect due to enriched alkaline functional groups and soluble Na⁺.⁵³ BC-HCl had the weakest effect due to acid-induced neutralization and alkali metal leaching. For BC-KMnO₄, oxidative modification selectively enhanced O-containing functional groups but also partially consumed BC alkalinity, leading to a moderate pH increase.⁵⁴ Under Cd²⁺ stress, the pH-elevating efficacy of BCs progressively declined with increasing exogenous Cd²⁺ levels, with only BC-NaOH maintaining a significant effect even at high Cd²⁺. BC-HCl caused soil pH decline due to residual acidity and Cd²⁺ hydrolysis-induced acidification.⁵⁵ Notably, at the highest exogenous Cd²⁺ concentration (40 mg/kg), only BC-NaOH elevated soil pH; all other BCs (BC, BC-HCl, BC-KMnO₄) caused significant pH reductions. This trend arose because excessive Cd²⁺ depleted soil alkalinity and overwhelmed the buffering capacity of non-alkaline BCs.⁵² BC-NaOH’s stable pH regulation relied on alkaline cation release and persistent deprotonation of –OH and –COO^– groups, reducing Cd²⁺ solubility and induced precipitation.⁵⁶

In the absence of exogenous Cd²⁺, all BC treatments showed a “rise-then-fall” OM content pattern, reflecting short-term C pool activation and long-term stabilization. Initial OM increase resulted from improved soil conditions, enhanced root exudate adsorption, and stimulated microbial activity.⁵⁷ The subsequent decline in OM resulted from the microbial mineralization of labile BC fractions, including adsorbed small-molecule organic compounds adsorbed on BC surfaces.⁵⁸ However, a high proportion of OM was stabilized by physical protection within BC’s porous structure and chemical binding to BC surface functional groups.⁵⁹ Among the treatments, BC-KMnO₄ exhibited the highest peak OM content, attributed to its increased surface –COOH and –OH groups, which enhanced the adsorption of soil organic colloids.⁶⁰ BC-NaOH also contributed to OM retention through two pathways. One was the strengthened electrostatic adsorption of organic colloids, which reduced microbial degradation.⁴⁵ The other was the provision of microhabitats for beneficial microbes, indirectly promoting OM synthesis.⁶¹ Additionally, the expanded macropore network of BC-NaOH physically shielded OM from rapid microbial decomposition.⁶² Under Cd²⁺ stress, OM magnitude decreased with rising Cd²⁺ levels, reflecting heavy metal inhibition of the soil C cycle.⁶³ Notably, low-dose (10 mg/kg) Cd²⁺ apparently induced a “mild stress” stimulation, promoting the secretion of polymers (e.g., polysaccharides) by microorganisms.⁶⁴ Moderate Cd²⁺ suppressed OM gain further, and at severe exogenous Cd²⁺ concentration of 40 mg/kg, only BC-NaOH and BC-KMnO₄ maintained significant OM retention, while BC and BC-HCl lost efficacy, highlighting their superior Cd²⁺-binding capacity.

Effects of unmodified and chemically modified BCs on HS composition under Cd²⁺ stress

In the absence of exogenous Cd²⁺ addition, all BC treatments facilitated the accumulation or transformation of HS fractions, though their efficacy varied significantly by modification type. BC-KMnO₄ exerted the most pronounced effects on the accumulation of C_WSS, C_HA, and C_Hu, followed by BC-NaOH; BC and BC-HCl exhibited weaker regulatory effects on these HS components. This gradient in efficacy highlighted that chemical modification enhanced BC’s capacity to regulate soil C cycling. Mechanistically, oxidative modification by KMnO₄ generated abundant surface –OH and –COOH groups, which enabled two key processes driving HS dynamics: acted as “C source carriers” and “microbial attachment sites” to support microbial growth and the secretion of extracellular polymeric substances that directly advance C_WSS synthesis^65,66; on the other hand, accelerating the transformation of native labile soil C into stable HS fractions, thereby promoting the accumulation of C_HA and C_Hu.⁶⁷ Notably, all modified BCs reduced the content of C_HE—compared to unmodified BC. This observation confirmed the “C stabilization” hypothesis, which posited that BC directed the transformation of labile C_HE into more stable HS fractions, thereby enhancing long-term soil C sequestration.⁶⁷

At low exogenous Cd²⁺ (10 mg/kg), Cd²⁺ stimulated the secretion of extracellular polymeric substances. In combination with Cd²⁺ immobilization by BCs (reducing toxicity), it promoted significant C_HA accumulation, especially in BC-KMnO₄ and BC-NaOH treatments.^65,66 At 20 mg/kg moderate exogenous concentrations of Cd²⁺, microbial activity was significantly suppressed, which drastically slowed C_WSS production. Notably, only BC-KMnO₄ and BC-NaOH maintained significant C_HA accumulation under this stress level. This concentration also marked the first occurrence of C_Hu degradation. Modified BCs—especially those with high surface negative charge and abundant O-containing functional groups (BC-KMnO₄, BC-NaOH)—outperformed BC and BC-HCl in stabilizing C_Hu.⁴⁵ At severe exogenous Cd²⁺ stress (40 mg/kg), severe inhibition of soil microbial activity and saturation of BC’s Cd²⁺ sorption sites led to the widespread degradation of the HS pool.⁶⁸ Notably, BC-NaOH retained a unique “stress resistance advantage,” attributable to its robust Cd²⁺ immobilization capacity and alkaline buffering effect—which alleviated Cd²⁺-induced soil acidification and thereby preserved higher HS stability.⁶⁹

Effects of unmodified and chemically modified BCs on HA molecular structure under Cd²⁺ stress

Regarding ΔlogK, in the absence of exogenous Cd²⁺, all BC treatments reduced ΔlogK relative to BC treatment, attributed to non-covalent interactions (e.g., π–π stacking and hydrogen bonding) between humification precursors and BC’s aromatic domains or polar functional groups, which promoted HA aggregation and structural stability.⁷⁰ BC-KMnO₄ and BC-NaOH induced the most pronounced HA condensation: BC-KMnO₄ facilitated aromatic ring polymerization via its strong oxidative activity, while BC-NaOH promoted HA aggregation through metal cation bridging.⁵⁶ BC-HCl had the weakest effect due to acid-induced alkaline metal loss.^71,72 At low exogenous Cd²⁺ (10 mg/kg), ΔlogK was further reduced in BC-KMnO₄ and BC-NaOH treatments. The mild Cd²⁺ stimulated the microbial secretion of extracellular polymeric substances, which acted as additional humification precursors, while BCs mitigated Cd toxicity to sustain humification.⁷³ At 20 mg/kg, exogenous concentrations of Cd²⁺, BC, and BC-HCl exhibited minimal regulatory influence, as Cd²⁺ began to suppress microbial humification.⁶⁸ At severe exogenous Cd²⁺ stress (40 mg/kg), BC and BC-HCl caused ΔlogK increases, indicating that HA had undergone depolymerization due to microbial inhibition and disrupted colloidal aggregation.⁶⁸ BC-KMnO4 and BC-NaOH still reduced ΔlogK, with BC-NaOH showing the strongest resistance via stable alkaline buffering.^53,69 At Cd²⁺ concentrations ≤20 mg/kg, all BC treatments significantly increased the C_HA/C_FA ratio. This enhancement was likely driven by BC-borne biomolecules (aliphatic compounds, polysaccharides) that served as labile C sources for humifying microbes, accelerating the transformation of FA (labile) to HA (stable).⁷⁴ BC-KMnO₄’s superiority was linked to its unique oxygen- and manganese-containing groups.⁷⁵ Under exogenous 40 mg/kg Cd²⁺ stress, C_HA/C_FA dropped for BC and BC-HCl, but increased marginally for BC-KMnO₄ (6.24%) and more substantially for BC-NaOH (20.0%), due to BC-KMnO₄’s MnO_X-related pore limitation and BC-NaOH’s stable alkalinity and macropore network.^29,54,76

Effects of unmodified and chemically modified BCs on Cd²⁺ concentrations, functional groups, and atomic molar of HA

Chemical modification of BCs markedly enhanced its capacity to modulate HA molecular structure. The strong oxidizing capacity of KMnO₄’s induced the cleavage of the BC’s C skeleton and generated abundant –OH and –COOH groups. These groups not only promoted the oxidative transformation of HA precursors but also increased the number of available adsorption sites.⁷¹ Alkaline etching by NaOH exposed latent O-containing functional groups within the BC, providing additional –OH and –COOH donors.²⁸ HCl modification exerted the weakest regulatory effect on HA structure. This was attributed to mild acidic conditions during modification, suppressing the deprotonation of –OH groups and limiting the formation of negatively charged –COO^- sites.⁴⁸ At Cd²⁺ levels (0 and 10 mg/kg), all BC treatments significantly enhanced the activity of –OH and –COOH groups in HA. As Cd²⁺ concentration increased (20 mg/kg), this enhancing effect diminished. Only BC-KMnO₄ retains substantial capacity to activate HA functional groups, attributed to its redox-active MnO_x layer.³⁰ Under exogenous 40 mg/kg Cd²⁺ stress, BC-KMnO₄ and BC-NaOH effectively alleviated the “Cd²⁺-HA complex-induced shielding” of –OH groups. In contrast, BC and BC-HCl exhibited reduced –OH group activity, as excessive Cd²⁺ formed stable Cd²⁺-HA complexes.⁷⁷ At Cd²⁺ (0, 10, and 20 mg/kg), (a+b+c)/e ratio decreased significantly in all BC treatments, indicating accelerated HA oxidation and aliphatic C breakdown. This was driven by mild Cd²⁺-induced stimulation of microbial activity, which was sustained by BC-mediated Cd²⁺ immobilization.⁶⁴ At high exogenous Cd²⁺ (40 mg/kg), the (a+b+c)/e ratio of BC increased, indicating suppressed HA oxidation and persistence of recalcitrant aliphatic structures, as severe Cd²⁺ stress inhibited microbial degradative enzymes.⁶⁸ BC-KMnO₄ and BC-NaOH—partially alleviated this inhibition. By efficiently immobilizing Cd²⁺ (reducing its bioavailability to microbes), these BCs sustained HA oxidation and promoted the degradation of less stable aliphatic C forms.⁷⁴ At Cd2+ levels (0 and 10 mg/kg), BC-NaOH and BC-KMnO₄ most strongly promoted HA aromatization via enhanced aromatic condensation.^75,78 At high exogenous Cd²⁺ (40 mg/kg), BC-KMnO₄ via MnO_x-mediated redox stabilization, and BC-NaOH via alkaline buffering to maintain HA’s aromatic domain integrity.^75,79

Atomic molar ratios in HA were influenced by the combined effect of Cd²⁺ and BC modification. Under low-to-moderate Cd²⁺ (≤20 mg/kg), all BC treatments reduced the HA’s H/C ratio, with BC-KMnO₄ exhibiting the strongest effect. This confirmed enhanced HA aromatization, driven by the BC-facilitated oxidation of aliphatic C.³⁵ At high exogenous Cd²⁺ (40 mg/kg), only BC-NaOH and BC-KMnO₄ maintained significant H/C ratio reductions, while BC and BC-HCl disrupted aromatic condensation via Cd²⁺ complexation.^80,81 BC-KMnO₄ most effectively raised the C/N ratio at lower Cd²⁺, by enhanced microbial activity and increased microbial N utilization in HA.⁸² However, at high exogenous Cd²⁺ (40 mg/kg), this effect was substantially diminished: This was attributed to severe Cd²⁺ suppression of microbial N mineralization and weakened BC regulatory capacity.⁸³ Under low-to-medium Cd²⁺, BC-NaOH and BC-KMnO₄ were most effective at raising O/C ratio, due to enhancements in both BC surface group content and HA oxidative modification.^53,75 Even at high exogenous Cd²⁺ (40 mg/kg), BC-KMnO₄ maintained the highest O/C ratio increment, demonstrating enduring oxidative functionality in highly contaminated soil. This was attributed to its MnO_x layer, which sustains redox cycling (Mn⁴⁺ → Mn³⁺ → Mn²⁺) to oxidize labile HA precursors into O-rich, structurally stable HA. This redox-driven oxidation compensated for the Cd²⁺-induced inhibition of microbial oxidative activity, ensuring continuous HA maturation.

Comprehensive assessment of BC and chemically modified BCs for Cd²⁺ immobilization and soil quality improvement in Cd-contaminated paddy soils

Key quality indicators (OM, BC pore volume, C_WSS, C_HA, and plant health metrics Cd²⁺ concentrations in rice leaf, stem, grain, and root) exerted a consistent and dominant influence on the PC1, explaining 73.32–79.88% of the total variance across all Cd²⁺ treatments (Figure 9). This indicated that these parameters were the most critical drivers of BC-mediated improvements in Cd-contaminated paddy soils. Under exogenous Cd²⁺ concentrations ≤20 mg/kg, BC-KMnO₄ outperformed other treatments by maximizing two core benefits: (1) soil quality improvements, including enhanced OM sequestration, C_WSS accumulation, and C_HA enrichment; and (2) optimized soil C cycling (evidenced by elevated C_HA/C_FA ratios and reduced HA ΔlogK values). These outcomes confirmed BC-KMnO₄ as the optimal amendment under mild-to-moderate Cd²⁺ stress. This was attributed to its unique combination of redox-active MnO_x sites (for Cd²⁺ immobilization) and abundant oxygen-containing functional groups.^30,60 In contrast, under severe exogenous Cd²⁺ stress (40 mg/kg), the primary criteria driving treatment differentiation shifted from PC1 to PC2. On PC2, soil pH, BC pore size, and HA structural ratios emerged as key differentiators. Here, BC-NaOH surpassed BC-KMnO₄ in maintaining critical soil traits. This was likely through two synergistic mechanisms: (1) enhanced the regulation of OM decomposition⁶² and (2) promotion of stable metal-organo complexes, facilitated by its high surface negative charge and alkaline buffering capacity, enabling it to maintain the stability of soil physical and chemical properties under severe Cd²⁺ pollution.^34,53

Economic viability and practical application prospects

Experimental results confirmed that key soil physicochemical properties (pH, OM content, and HS fractions) remained stable after BC amendment, verifying no secondary contamination risks of modified BCs (BC-HCl, BC-NaOH, and BC-KMnO₄), which met the basic safety criteria for agricultural soil remediation. First, rice husk, the feedstock for BC, was a non-toxic, biodegradable agricultural by-product with abundant availability and no exogenous pollutants, avoiding undesirable substance input into soil. Second, the modification process was strictly controllable: repeated rinsing of BC until the filtrate reached neutrality ensured the efficient removal of residual chemical modifiers and potential reaction by-products, with concentrations well below regulatory limits in soil amendment standards. Meanwhile, the modified BCs significantly enhanced Cd²⁺ immobilization efficiency and soil environmental quality, ensuring ecological safety in food-producing paddy systems. Economically, chemical modifiers (HCl, NaOH, and KMnO₄) could be recovered from pre-treated industrial waste (e.g., effluents generated by chemical synthesis or metal processing), realizing waste valorization while enhancing BC’s Cd²⁺ remediation efficiency.

Limitations of the study

A limitation of this study was that it did not incorporate the regulatory role of microorganisms in nutrient turnover and BC mediated Cd²⁺ immobilization, which warranted further investigation to comprehensively clarify the holistic remediation mechanisms in Cd-contaminated paddy soils.

Resource availability

Lead contact

Requests for further information and resources should be directed to and will be fulfilled by the lead contact, Shuai Wang (wangshuai@jlnku.edu.cn).

Materials availability

This study did not generate new unique reagents.

Data and code availability

This paper does not report original code. All data associated with this study are present in the paper or the supplemental information. Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

Acknowledgments

This study was funded by the Science and Technology Project of the Education Department of Jilin Province, China (grant no. JJKH20240504HT).

Author contributions

Conceptualization, K.L. and S.W.; methodology, D.D. and J.W.; visualization, J.G., M.W., and Q.H.; data curation, P.L., M.Y., and J.P.; writing – original draft, K.L. and S.W.; writing—review and editing, K.L. and S.W.; funding acquisition, K.L.; resources, S.W., D.D., and J.W.; supervision, S.W. All authors have read and agreed to the published version of the article.

Declaration of interests

The authors declare no competing interests.

STAR★Methods

Methods details

Experimental design

Surface soil samples (0–20 cm depth) were collected from representative rice paddies in the central region of Jilin Province, China. Upon collection, the soil was air-dried at ambient temperature, gently ground to disperse aggregates, and passed through a 2 mm nylon sieve to remove coarse debris (e.g., plant residues, stones) prior to subsequent analyzes and experiments. The basic physicochemical properties of the soil were determined as follows: pH 6.04 ± 0.08, OM content 16.5 ± 0.8 g/kg, available potassium 288.3 ± 3.2 mg/kg, and background Cd²⁺ content 0.28 ± 0.05 mg/kg.

Raw rice husks were used as the feedstock for BC preparation via oxygen-limited pyrolysis. Specifically, raw rice husks were placed in a muffle furnace and pyrolyzed at 500°C for 2 h under anoxic conditions (by purging with nitrogen gas at a flow rate of 100 mL/min to minimize oxidative degradation). The resulting BC was cooled to room temperature, ground in a mortar and pestle, and sieved through a 0.15 mm nylon sieve. This unmodified biochar was designated as BC (I). To enhance its reactivity toward Cd²⁺, three types of chemically modified BCs were synthesized, with detailed procedures below: HCl-modified biochar (BC-HCl, Ⅱ): BC was mixed with 1.0 mol/L HCl solution at a solid-to-liquid ratio (m/v) of 1:20. The mixture was continuously shaken at 150 rpm for 24 h at 25°C in a constant-temperature shaker. After oscillation, the solid phase was separated by vacuum filtration and rinsed repeatedly with deionized water until the filtrate reached a neutral pH (pH 6.8–7.2). The washed product was dried at 60°C for 12 h to a constant weight, then ground and sieved through a 0.15 mm sieve to obtain BC-HCl. NaOH-modified biochar (BC-NaOH, Ⅲ): BC-NaOH was prepared following the same procedure as BC-HCl, except that a 1.0 mol/L NaOH solution was used instead of HCl (solid-to-liquid ratio 1:20, shaking at 150 rpm for 24 h at 25°C, followed by neutral rinsing and drying at 60°C). KMnO₄-modified biochar (BC-KMnO₄, IV): BC was mixed with a 0.1 mol/L KMnO₄ solution at a solid-to-liquid ratio (m/v) of 1:15. The mixture was incubated at 25°C in the dark for 6 h with continuous stirring at 100 rpm to ensure sufficient contact between the oxidant and the BC surface, thereby enabling uniform oxidative modification and minimizing the risk of localized under-modification or over-modification. This approach also prevented the photodegradation of KMnO₄. After incubation, the solid was filtered, rinsed thoroughly with deionized water until the filtrate turned colorless (confirming complete removal of unreacted KMnO₄). The product was dried at 60°C for 12 h, then ground and sieved through a 0.15 mm sieve to yield BC-KMnO₄.

All chemicals used in this study were of analytical grade or agricultural standard and commercially available: cadmium nitrate tetrahydrate: Cd(NO₃)₂·4H₂O (analytical grade, 0.5 mol/L stock solution), urea (N ≥ 46%), potassium chloride (K₂O ≥ 60%), and calcium superphosphate (P₂O₅ ≥ 12%). Among these, urea, potassium chloride, and calcium superphosphate are commonly used in agricultural practices as fertilizers.

Plastic sorting boxes (0.6 m in length, 0.4 m in width, and 0.3 m in height) were each filled with 5.0 kg of the prepared soil, into which 250 g of each BC types (BC, BC-HCl, BC-NaOH, or BC-KMnO₄) was uniformly mixed into the soil, corresponding to an 5% application rate relative to the total soil mass. Concurrently, basal fertilizers were applied at the following rates: 2.32 g of urea, 6.67 g of calcium superphosphate, and 1.00 g of potassium chloride per box. During the tillering stage of rice, an additional topdressing was administered, consisting of 2.32 g of urea and 1.00 g of potassium chloride per box. This fertilization regime was equivalent to a total field-applied rate of 200 kg/ha (N), 80 kg/ha (P₂O₅), and 120 kg/ha (K₂O), consistent with local rice cultivation practices.

Four Cd²⁺ concentration levels (0, 10, 20, and 40 mg/kg), designated as Cd0, Cd10, Cd20, and Cd40, were established by spiking the soil with Cd(NO₃)₂·4H₂O.⁸⁴ The experiment followed a completely factorial design with four BC types and four Cd²⁺ levels, yielding 16 treatment combinations. For each treatment, four healthy rice seedlings were transplanted into one plastic sorting box, with three replicate boxes per treatment.

Rice plants (Oryza sativa L. ssp. japonica cv. Zhongkefa 5) were grown in a growth chamber with controlled environmental conditions, equipped with supplemental LED lighting (Light intensity: 35,000 lx; Light duration: 12 h), for a total of 120 days (d). The growth parameters were maintained constant throughout the experiment: temperature (25 ± 1°C), relative humidity (70%). Periodic irrigation with deionized water was performed to maintain soil moisture at an optimal level for rice growth. Soil samples were collected from each replicate box at 0, 30, 60, 90, and 120 days. At each sampling event, a minimum of 200 g of soil was collected per box, with samples taken from three random positions within the box and homogenized to ensure representativeness. Upon completion of the cultivation period, entire rice plants were harvested and separated into roots, stems, leaves, and grains. All plant organs were dried at 105°C for 2 h, followed by drying at 85°C to constant weight, then ground into a fine powder using a stainless-steel grinder and passed through a 0.10 mm nylon sieve. The ground samples were stored in sealed polyethylene containers at room temperature until subsequent chemical analysis.

Quantification and statistical analysis

Determination of chemical properties

Surface morphology and oxygen-containing functional groups analysis: The external surface morphologies of BC, BC-HCl, BC-NaOH, and BC-KMnO₄ were characterized via SEM (Model SS-150 days-ST, Shanshi Instrument Co., Ltd., Shenzhen, China) at an accelerating voltage of 15 kV and a magnification range of 500–5000×. This magnification range was selected to capture both macrostructural features (e.g., pore distribution) and microstructural details (e.g., surface deposits on BC-KMnO₄). SEM micrographs were acquired to support subsequent qualitative analysis.The specific surface area (SSA), porosity and pore size of all BC samples were determined by Static Volumetric Method using the Surface Area Porosity Analyzer (model BSD-PS2, Best Instrument Technology Co., Ltd., Beijing, China). Fourier transform infrared (FTIR) spectroscopy (Model FTIR-850, Gangdong Science and Technology Development Co., Ltd., Tianjin, China) was employed to qualitatively identify the composition of oxygen-containing functional groups in all BC samples.

Determination of Cd²⁺ content in rice organs

For each rice organ (root, stem, leaf, and grain), 0.5 g of dry sample was accurately weighed into a digestion tube, to which 10 mL of a HNO₃-HClO₄ mixture (volume ratio 9:1) was added. The mixture was allowed to soak for 12 h to facilitate pre-digestion. Digestion was then performed by heating the tubes on a hot plate, continuing until the solution became colorless, transparent, or pale yellow (indicating complete digestion). After cooling to room temperature, each digestate was transferred to a 25 mL volumetric flask and diluted to volume with ultrapure water. The Cd²⁺ concentration in the diluted digestates was determined using atomic absorption spectrometry with a graphite furnace (AAS-GF, Model TAS-990 Super AFG, Puxi General Instrument Co., Ltd., Beijing, China).

Extraction of HS components and determination of their C contents

HS fractions were isolated using a modified Kumada extraction method, with the detailed procedure described below: A 5.0 g soil sample was accurately weighed into a 100 mL polyethylene centrifuge tube, and 30 mL of distilled water was added. After thorough mixing, the suspension was subjected to extraction in a water bath shaker at 70°C for 1 h, followed by centrifugation at 3,500 rpm for 15 min. The supernatant was carefully transferred to a 50 mL volumetric flask. The residual soil in the centrifuge tube was subjected to a second extraction with 20 mL of distilled water, and the centrifugation step (3,500 rpm, 15 min) was repeated. The supernatants from the two extractions were combined and diluted to the 50 mL mark with distilled water; this fraction was designated as the WSS. For the extraction of HE fraction, distilled water was replaced with a mixed solution containing 0.1 mol/L NaOH and 0.1 mol/L Na₂P₂O₇·H₂O. The above extraction-centrifugation procedure was duplicated, and the resulting supernatant was collected as the HE fraction. The remaining solid residue after HE extraction was dried at 60°C and sieved through a 0.15 mm nylon sieve to obtain the Hu fraction.

To separate FA and HA from the HE fraction, a 30 mL aliquot of the HE solution was acidified to pH 1.0–1.5 with 0.5 mol/L H₂SO₄. The acidified solution was incubated in a 70°C water bath for 1.5 h to promote precipitation of HA, then removed from the bath and allowed to equilibrate overnight at room temperature. On the following day, the solution containing the precipitate was filtered, and the filtrate was transferred to a 50 mL volumetric flask, diluted to volume with ultrapure water, and defined as the FA fraction. The precipitate retained on the filter paper (identified as humic acid, HA) was washed with acid to remove impurities, then redissolved in warm 0.05 mol/L NaOH. The resulting solution was diluted to the marked volume of a volumetric flask to prepare the HA stock solution for subsequent analysis.

The C_WSS, C_HE, C_HA, and C_Hu were determined using the external heating potassium dichromate oxidation method. The C content of FA (C_FA) was calculated as the difference between C_HE and C_HA (C_FA = C_HE-C_HA), and the C_HA/C_FA ratio was computed accordingly. Additionally, the ΔlogK coefficient of the HA alkaline solution was analyzed via a UV–visible spectrophotometer (Model TU-1900, Purkinje General Instrument Co., Ltd., Beijing, China) by measuring the absorbance at 400 nm and 600 nm; the ΔlogK value was calculated using the formula: ΔlogK = lgabs400-lgabs600. The functional groups of HA were characterized by FTIR, and the elemental composition of HA was analyzed using a PerkinElmer CHNS/O Analyzer (Model 2400, PerkinElmer Inc., Waltham, MA, USA).

Statistical analyses

All experimental data were processed and statistically analyzed using Origin 9.0 (OriginLab Corporation, Northampton, MA, USA) and Microsoft Excel 2017 (Microsoft Corporation, Redmond, WA, USA). One-way analysis of variance (ANOVA) was conducted with SPSS Statistics 20.0 software (IBM Corporation, Armonk, NY, USA) to evaluate the significance of differences among treatment groups, with the significance level set at p < 0.05. Post hoc multiple comparisons of means were performed via least significant difference (LSD) test. All quantitative data were expressed as the mean ± standard deviation (SD) of three biological replicates. For all bar charts generated to visualize results, error bars represent the SD values of the three replicates, thereby offering a clear illustration of data variability.