Nanomaterials Reduce Cadmium Bioavailability in Paddy Soils Through Redox-Driven Immobilization and Microbial Dynamics

Abstract

Cadmium (Cd) mobilization in paddy soils during redox fluctuations poses significant risks to rice safety. This study investigated the efficacy of nano-calcium carbonate (NCC), nano-hydroxyapatite (NHAP), and their composite (C+P) in immobilizing Cd under simulated flooding–drainage cycles. Soil treatments (0.5% and 1.0% w/w) were subjected to 40 day anaerobic and 20 day aerobic incubation. The results demonstrated that NCC and C+P elevated the soil pH by 1.35–1.39 and 0.72–1.01 units during the anaerobic and aerobic phases, respectively. These amendments suppressed Fe(II) and Mn(II) release by 41–75%, correlating with reduced Cd bioavailability. While nanomaterials minimally influenced Cd speciation during flooding, aerobic conditions triggered a marked shift: residual Cd fractions increased by 80.8–116.4% under NCC, driven by CdCO₃ precipitation and phosphate complexation. Cd release rates decreased by 53.6–66.8% in NCC and C+P treatments during oxidation. Microbial analysis revealed diminished bacterial diversity but enriched Firmicutes (up to 58.9%), which positively correlated with pH and residual Cd. Redundancy analysis identified pH and Fe/Mn dynamics as key regulators of the microbial community structure. NCC emerged as the most effective amendment. This study highlights the potential of NCC-based strategies for mitigating Cd risks in acidic paddy soils, particularly during post-flooding drainage.

1. Introduction

Anthropogenic activities have released significant quantities of heavy metals into agricultural land, posing a threat to soil health and the safe production of crops [1,2]. Cadmium (Cd), a non-essential element in biological systems, often exists in compound form in nature and is one of the most hazardous and toxic metals in ecosystems [3,4]. For instance, long-term consumption of rice contaminated with Cd can induce pathological changes in multiple human organs, including the kidneys, bones, and lungs [5]. The biological toxicity of Cd in soil largely depends on its speciation [6,7,8], with redox states being significantly more toxic than adsorption and precipitation states in terms of bioavailability [9]. However, the forms in which Cd exists in the soil are determined by the environmental conditions, such as pH [10]. Therefore, altering the physicochemical properties of the soil is an effective strategy to reduce the concentration of Cd in agricultural soils, particularly in rice paddy fields [11,12].

During the rice growing period, paddy soil is characterized by periodic flooding and drainage. Continuous flooding occurs until approximately 20 days before harvest, followed by drainage to control excessive tillering [13]. This flooding and drainage process can lead to a significant increase in the mobilization of Cd in the soil [14], subsequently raising the Cd concentration in rice grains. Notably, 80% of the Cd in rice grains accumulates during the grain-filling stage [15]. Therefore, to reduce the Cd concentration in rice grains, increasing soil pH values during the oxidation stage is an effective approach to enhance Cd immobilization in the soil.

The use of nanomaterials has emerged as a promising area in environmental remediation due to their unique size-dependent properties [16], which confer enhanced reactivity and sorption capabilities compared to conventional materials [17]. Nanomaterials stabilize heavy metals through direct interactions or by indirectly altering soil properties, thereby immobilizing their transformation and bioavailability [18]. These materials are highly effective in controlling plant diseases, remediating soil, and promoting plant growth [19,20,21].

Nano-calcium carbonate (NCC) and nano-hydroxyapatite (NHAP) are ultrafine solid materials with various advantages, including low cost, strong reinforcement, and ease of processing. Both materials exhibit unique properties, such as the size effect, large surface area, and macro quantum tunneling effect, which are not present in non-nanomaterials [22,23]. Additionally, due to the absence of adjacent atoms and the presence of numerous dangling bonds, NCC possesses high chemical energy. The increase in the number of dangling bonds facilitates atomic binding, thereby enhancing stability. For instance, the adsorption capacity of NCC for Cd is as high as 2207.4 mg g⁻¹, significantly higher than bulk CaCO₃ (500–800 mg g⁻¹) due to its enhanced surface reactivity and quantum tunneling effects [24]. Meanwhile, NHAP can reduce the exchangeable Cd fractions by approximately 10% and enhance the residual Cd fractions by 34% to 50%, surpassing bulk HAP [25]. Therefore, these two nanomaterials hold significant potential for immobilizing Cd in soils. Numerous studies have focused on reducing the Cd concentration in rice grains by applying passivation materials to the soil; however, they have not thoroughly examined the critical stage of rice filling and the primary factors that influence Cd activation during the reduction–oxidation stage. Thus, we hypothesized that the nano-materials NCC and NHAP can effectively reduce the bioavailable fractions during the oxidation stage (rice filling stage).

Here, we simulated the flooding and drainage processes of typical soil contaminated with Cd. We systematically investigated the effects of adding various types, amounts, and composites of nanomaterials on the release rate of Cd, as well as on pH, redox potential (Eh), major dissolved substances, available Cd, and microbial community diversity in the soil. The specific aims were: (1) to evaluate the effects of NCC and NHAP on reducing soil Cd bioavailability and release rate; (2) to investigate the factors and mechanisms by which the addition of NCC and NHAP influences soil Cd transfer during the oxidation–reduction process in paddy soil; and (3) to explore the potential for improving soil quality by enhancing soil microbial communities.

2. Materials and Methods

2.1. Preparation of Nanomaterials

NCC was produced using the carbonation–precipitation technique [26]. Centrifugation was conducted on a 1 L solution of Ca(OH)₂ (0.5 M) while passing 20 mL/min of a 40% CO₂ volume concentration at 25 °C for 30 min. A pH meter (PB-21, Sartorius, Germany) was used to continuously monitor the pH of the suspension during the mixing process. The suspension and intermediate precipitate were promptly washed with ethanol to inhibit crystal growth. The washed suspension was then filtered and dried at 80 °C.

The preparation method for NHAP involved the solution co-precipitation technique [27]. Initially, calcium carbonate nanopowder was added to deionized water to create a calcium carbonate nanosuspension (0.01–0.1 mol/L) using ultrasonic vibration. A sodium phosphate suspension (0.01 mol/L) was prepared by dissolving sodium phosphate in deionized water (0.1 mol/L) and then diluting it to 0.01 mol/L. The pH of the sodium phosphate suspension was adjusted to a range of 9.5–10.5 using a sodium hydroxide solution. Subsequently, the calcium carbonate nanosuspension was added dropwise to the sodium phosphate solution at a rate of 1 mL/min. The reaction was carried out using an electrically heated magnetic stirrer at a temperature range of 120–150 °C, with a stirring speed set at 150 rpm for 8 h. Finally, the resulting product was filtered, washed with deionized water, and dried to obtain NHAP powder.

The produced NCC and NHAP are spherical in shape, with particle sizes ranging from 30 to 80 nm and an average diameter of 50 nm. Mineral morphology was analyzed using a scanning electron microscope (SEM) and X-ray diffraction (XRD). The detailed method is presented in Test S1. The SEM images are shown in Figure S1 and the XRD of NCC is shown in Figure S2, while the XRD of NHAP is illustrated in our previous study [28]. XRD results of NCC revealed that the peak characteristics appear at 2θ = 29.4°, 35.9°, 43.1°, 47.4°, and 48.5°, which are consistent with the calcite pattern described in ASTM card number 1, primarily composed of calcium carbonate powder. This morphology facilitates a larger specific surface area, thereby creating more adsorption sites for heavy metal ions. The surface exhibits a rough and irregular porous structure, characterized by uniformly distributed granular objects. The NCC suspension (1% w/w) had an initial pH of 9.2 vs. NHAP (7.8), confirming its stronger alkalinity.

2.2. Soil Incubation Experiments

The test soil was sampled from a paddy field (0–20 cm) in Xinyu City, China (114°97′ E, 27°78′ N). The paddy site was contaminated with Cd (0.58 mg kg⁻¹) due to long-term mining activities. Since air-drying soil does not significantly impact the microbial community composition or structure [29], air-dried soil was used for analysis in this study. After air-drying, the sampled soil was passed through a 2 mm mesh screen. The physicochemical properties of the soil are detailed in the Supplemental Information (Table S1).

The incubation experiment simulated the flooding and drainage processes of typical Cd-contaminated soil in September to October 2023, following the modified microcosm method described in a previous study [30] and also detailed in Text S2. We established seven treatments: NCC, NHAP, and a 1:1 ratio mixture of NCC and NHAP (C+P) at two application mass ratios (material: soil) of 0.5% (w/w) and 1% (w/w), along with a control treatment. The 0.5–1.0% application rate (5–10 t/ha) aligns with conventional lime usage, but NCC’s efficiency may lower field costs [11].

The materials were mixed homogeneously with the soils and labeled as 0.5% NCC, 1.0% NCC, 0.5% NHAP, 1.0% NHAP, 0.5% C+P, and 1.0% C+P, respectively. Each treatment had 30 replicates. Specifically, 30 g portions of the collected soils mixed with nanomaterials were placed into 100 milliliter vials, to which 60 mL of deionized water was added. The vials were initially incubated anaerobically for 40 days (referred to as the “reducing phase” (R)), followed by an additional 20 days of aerobic incubation (referred to as the “oxidation phase” (O)).

2.3. Sampling and Analysis

The redox potential (Eh), pH, soil solution Cd, and other major ions in the soil solution were analyzed during the R phase and O phase. At various incubation periods within the R phase (0, 1, 3, 5, 15, and 40 days; hereafter referred to as R0, R1, R3, R5, R15, and R40) and the O phase (1, 2, 5, 10, and 20 days; hereafter referred to as O1, O2, O5, O10, and O20), three vials for each treatment were destructively collected for solution analysis. The soil Eh, soil solution pH, dissolved Cd, and dissolved iron (Fe) and manganese (Mn) were measured, as detailed in Test S1. Unless stated otherwise, all operations were performed in a glovebox under an argon atmosphere. The soil samples were collected after centrifugation and divided into two subsamples. The distribution of Cd fractions was determined using the Community Bureau of Reference (BCR) continuous extraction method for the soil samples collected at R3, R40, O2, O10, and O20. The BCR continuous extraction yielded the acid fraction, reducible fraction, oxidizable fraction, and residual fraction of Cd in the soils (Table S2). Mineral morphology was analyzed using a scanning electron microscope (SEM) and X-ray diffraction (XRD). XRD analyses were conducted using a Shimadzu XRD-6100 diffractometer, and data was collected in the 2θ angular range of 10–80° at a scanning rate of 1.0° min−1 with a step of 0.02°. The microstructure and surface morphology composites were observed with a Hitachi S4800 SEM. The XRD results of NCC revealed that the peak characteristics appear at 2θ = 29.4°, 35.9°, 43.1°, 47.4°, and 48.5°, which are consistent with the calcite pattern described in ASTM card number 1, primarily composed of calcium carbonate powder.

Another set of fresh soil samples (R40, O20) was stored in an ultra-low temperature freezer at −80 °C to investigate the changes in diversity and abundance of soil microbial communities following the addition of nanomaterials. The composition of these microbial communities was analyzed using 16S rRNA sequencing after the introduction of nanomaterials at various stages of water management. The Illumina MiSeq system was employed to sequence 21 samples, yielding a total of 749,810 raw sequences. High-quality bacterial sequences were obtained by utilizing TrimMomatic software (version 0.36) to filter the raw sequencing data for quality control. These sequences were subsequently clustered into 14,941 operational taxonomic units (OTUs) using a 97% similarity threshold (Figure S3). The saturation of the rarefaction curve indicates that the generated bacterial sequences accurately represent the bacterial community under investigation (Figure S1). We calculated alpha (α) and beta (β) diversity from the OTU table and constructed a rarefaction curve to compare OTU richness. The rarefaction curve analysis and the Chao1 index were employed to assess soil diversity, while the Shannon diversity index was used to evaluate microbial diversity. Principal coordinate analysis (PCoA) of the soil bacterial communities, based on the Bray–Curtis distances, was conducted to visualize the differences among the various groups. We focused on the microflora at the phylum taxonomic level to analyze the composition and abundance of microbial communities. Redundancy analysis (RDA) was performed to examine the influences of changes in soil properties and Cd fractions on the composition of the microbial communities at the phylum level.

2.4. Statistical Analyses

The Cd release rate coefficient (k) was calculated:

$$\mathrm{k}=({\mathrm{C}}_{\mathrm{t}}-{\mathrm{C}}_{\mathrm{0}})/\mathrm{d}$$

where C_t and C₀ represent Cd concentrations at time t and initial state, respectively; d represents days. All results presented in this study are expressed as the mean ± standard deviation (SD). Statistical analyses were conducted using one-way analysis of variance (ANOVA), followed by Duncan’s test to identify significant differences between treatment means (p < 0.05 and p < 0.01). Graphs were generated using Origin 9.1 software (OriginLab Corporation, Northampton, MA, USA) and GraphPad Prism 9.5.

3. Results and Discussion

3.1. Dynamics Changes in pH, Eh, and Soluble Fe and Mn Concentrations

The pH of the soil solution fluctuated during the incubation phase (Figure 1A). Soil addition of NCC showed a much higher increase of soil pH compared to the other treatments, because CO₃²⁻ in the soil solution can assume the H⁺ to produce HCO₃¹⁻ or H₂CO₃ [31,32]. Similarly, the pH values were elevated in the NHAP treatments, because NHAP can dissolve and releases PO₄³⁻ in the soil solution, which reacts with H⁺ to form HPO₄²⁻ or H₂PO₄⁻, leading to an increase in the pH value [33]. During the reducing phase, the pH of the soil solution gradually increased to near-neutral values as the duration of flooding increased. Compared to the control group, the 0.5% and 1.0% NCC treatments during the reducing phase raised the soil pH by 1.34 and 1.39 units, respectively. Similarly, the 0.5% and 1.0% NHAP treatments increased the soil pH by 1.16 and 1.25 units, while the 0.5% and 1.0% C+P treatments elevated the soil pH by 1.35 and 1.38 units, respectively (Figure 1A). These increases are likely due to the reaction of Fe(III) and Mn(IV)/Mn(III) oxides, as well as sulfates in the soil, with organic reducing substances. This reaction forms Fe(II), Mn(II), and S²⁻, which consume protons (H⁺) and lead to an increase in soil pH [15]. As the reducing conditions intensified, the continuous depletion of protons and the slowing of the CO₂ diffusion rate resulted in a gradual increase in soil pH [8,34,35]. The decrease in H⁺ concentration increases the adsorption sites on the surface of nanomaterials and promotes the fixation of Cd, which may contribute to the reduction in soil Cd bioavailability [36,37].

The pH initially increased to a maximum value of approximately 8.00 at the beginning of the re-oxidation phase, which has not been observed in previous soil flooding-drainage studies. We speculate that it may be caused by OH⁻ release caused by oxidation of reducing substances, and the specific mechanism behind it still needs to be further studied. Then, it gradually decreased to near neutrality. Soil drainage increased the oxygen content, promoting the re-oxidation of Fe(II) and S re-oxidization. Protons were produced during these processes, leading to a decrease in soil pH. Compared to the control group, the soil pH increased by 0.87 units with the 0.5% NCC treatment and by 1.01 units with the 1.0% NCC treatment during the oxidation phase. Similarly, the 0.5% and 1.0% NHAP treatments increased the soil pH by 0.29 and 0.61 units, while the 0.5% and 1.0% C+P treatments raised the soil pH by 0.72 and 0.83 units, respectively. In conclusion, both the reducing and oxidation phases demonstrated the most significant impact of NCC in increasing pH, likely due to the inherently higher pH of the NCC material itself.

In contrast to the changes in pH, the Eh values exhibited opposite trends (Figure 1B). Numerous studies have demonstrated that Eh significantly influences the bioavailability of Cd in soil [38,39,40]. The variations in Eh are determined by the ratio of oxidizers to reducers present in the environment [41]. The Eh value of the soil solution rapidly decreased to below 0 mV after 10 days during the reducing phase and subsequently dropped to its lowest range of −200 mV to −100 mV by day 40 of the same phase. This decline was attributed to the soil being submerged and in an anaerobic state, which leads to incomplete decomposition of organic matter. This incomplete decomposition generates organic reducing substances, creating a reducing environment in the soil and further resulting in a rapid decrease in its Eh value [42]. During the oxidation phase, the Eh quickly returned to the initial value or slightly below it within 20 days (Figure 1B).

Rapid increases in soluble Fe and Mn concentrations were observed within 15 days of the reducing phase (Figure 2), attributed to microbial reductive solubilization of Fe(III) and Mn(III/IV) oxides. Subsequently, the concentrations of soluble Fe and Mn began to decline slightly and eventually dropped significantly to, or slightly below, their initial values upon entering the oxidation phase. The activity of Cd varied cyclically with paddy soil water management, which is significantly correlated with the redox state of Fe oxides in the soil and various soil conditions (e.g., pH, humidity, microbial species), thereby greatly affecting Cd activities [43]. Compared to the control group, the amount of Fe(II) produced through the reductive dissolution of Fe and Mn oxides was noticeably lower in the treatment groups, particularly in the 0.5% NCC and 1.0% C+P treatments (Figure 2). This may be due to the treatment group significantly increasing the pH to neutral levels, leading to the absorption of soluble Fe(II) by excess Fe oxides, which generated mixed minerals of Fe(II) and Fe(III), such as siderite (FeCO₃), vivianite (Fe₃(PO₄)₂·8H₂O), goethite (α-FeOOH), hematite (α-Fe₂O₃), and magnetite (Fe₃O₄). This process resulted in the sequestration of Cd into the newly formed secondary iron minerals [44,45], which also explains the significant decrease in dissolved Cd content. The concentrations of Mn(II) underwent significant redox transformations due to paddy soil water management, similar to the changes observed in Fe(II). The concentrations of Mn(II) stimulate the growth of Fe oxides, accelerating the development of amorphous Fe oxides in the soil and resulting in the formation of Fe/Mn oxide plaques in the root system. The Mn nutritional status of the crop may influence Cd uptake [46].

3.2. Dynamics of Available Cd During Reduction/Oxidation Alternations

Heavy metal species in soil are commonly identified through sequential extraction methods [47]. In this study, BCR sequential extraction analysis was conducted on the amended soil to investigate metal immobilization and transformation. The results of continuous fractionated extraction using the BCR method indicate that as soil flooding increases, the percentages of acid-extractable and reducible fractions decrease, while the percentages of oxidizable and residual fractions increase across all treatments during the reduction stage (Figure 3). Soil flooding reduces the availability of Cd in acidic soils by causing these soils to transition to inactive fractions [48,49]. Soil pH is a significant factor influencing Cd bioavailability, and flooding induces a shift in Cd activity, which is associated with changes in soil pH. An elevated pH level facilitates the precipitation of Cd and increases the number of negatively charged adsorption sites, thereby promoting the adsorption of Cd from the soil. Consequently, the formation of carbonate and reducible Cd results in decreased availability [42,50]. At the end of the reduction phase (R40), there was no significant difference in Cd activity reduction among the treatment groups compared to the control group, suggesting that flooding plays a more critical role in the transformation of Cd within the residual fraction.

In the oxidation phase, the proportion of acid-extractable and reducible fractions in the control group gradually increased with soil aeration time. In the oxidation phase, the proportion of acid-extractable and reducible fractions in the control group gradually increased with the duration of soil aeration. This increase was accompanied by a decrease in the proportion of oxidizable and residual fractions. This finding aligns with a previous study [42], which suggested that the drainage process can induce hazardous changes in the activity level of Cd in acidic paddy soils, potentially affecting both organisms and the environment. Compared to the control group, the 0.5% NCC and 1.0% NCC treatments resulted in increases of 80.8% and 116.4%, respectively, in the proportion of residual fractions (Figure 3B,C). Similarly, the 0.5% NHAP and 1.0% NHAP treatments led to increases of 53.9% and 88.0% in the proportion of residual fractions (Figure 3D,E). Additionally, the 0.5% C+P and 1.0% C+P treatments resulted in increases of 56.8% and 70.2% in the residual fractions (Figure 3F,G). The proportion of residual fractions increased with the addition of nanomaterials, particularly in the treatments involving NCC. This phenomenon may be attributed to the dissolution of CaCO₃ on the surface of NCC, followed by the formation of CdCO₃ precipitates [51]. These precipitates are stabilized against Cd ions through adsorption. Furthermore, soluble phosphorus from NHAP precipitates adsorbs Cd, forming relatively stable Cd-containing phosphates with a unique crystal structure, which effectively transform Cd into a stable fraction [25,52].

3.3. Transfer Process of Cd Release from Soils

During the entire soil incubation phase, the concentration of dissolved Cd in each treatment exhibited a fluctuating decrease and then gradually stabilized by the 60th day of incubation (Figure 4A). In the initial phase of incubation, the Cd concentrations in the soil solution significantly decreased across all treatments. This reduction is likely attributed to the transformation of soluble Cd into more stable forms during the reducing phase, such as residual fractions (see Section 3.2). Following the rapid release of Cd, the concentration of dissolved Cd gradually decreased to nearly zero by Day 60. Furthermore, compared to the control, lower concentrations of dissolved Cd were observed in the 0.5% NCC, 1.0% NCC, 0.5% C+P, and 1.0% C+P treatments during the equilibrium period between Days 15 and 40 (Figure 4A), indicating that these treatments effectively immobilized soil Cd.

During the reducing phase of soil incubation, the Cd release rate was reduced by 68.9% and 71.8% in the 0.5% and 1.0% NCC treatments, respectively, and by 66.9% and 74.7% in the 0.5% and 1.0% C+P treatments compared to the control group. Lower immobilization effects were observed in the 0.5% and 1.0% NHAP treatments, which reduced the Cd release rate by 41.0% and 55.3%, respectively (Figure 4B). Overall, there were no significant differences between the effects of the NCC and C+P treatments. During the oxidation phase, the release rate of Cd in the control group significantly increased. Treatments with 0.5% NCC, 1.0% NCC, and 1.0% C+P reduced the Cd release rate by 53.6%, 66.8%, and 62.7%, respectively, compared to the control group (Figure 4B). Our results indicate that NCC treatment significantly decreased the rate of Cd release from paddy soil. NCC possesses a greater number of adsorption sites on its surface, and since Cd²⁺ has an ionic radius similar to that of Ca²⁺, NCC exhibits a higher affinity for Cd²⁺. These findings further confirmed that NCC has a superior capacity and strength to adsorb Cd²⁺ from soil solutions, thereby mitigating the risks of Cd contamination.

3.4. Soil Bacterial Communities and Functions

Soil microbial communities serve as indicators of changes in soil quality, while soil versatility and microbial diversity are limited in the presence of heavy metals [53]. The addition of nanomaterials has impacted the environment for soil bacteria, resulting in alterations to the diversity and composition of the bacterial community. In assessing microbial community abundance and diversity, the Chao1 (species richness) and Shannon diversity (species diversity) indices are widely recognized as optimal indicators [54]. During the reducing phase of soil incubation, the 1.0% NCC treatment resulted in a significant increase in both the Chao1 index and the Shannon index (Figure 5). However, no significant differences in these indices were observed among the other treatments. During the oxidation phase, bacterial community diversity was significantly reduced in the nanomaterial-added treatments compared to the control group. This phenomenon suggests that exposure to nanomaterials markedly decreased soil microbial community diversity, particularly during the drainage phase. This may be attributed to the fact that soil drainage enhances the contact between nanomaterials and biotic or abiotic molecules in the soil matrix, thereby altering their surface properties and potential toxicity to microbial cells [55,56]. Nonetheless, akin to other xenobiotics, the detrimental impact of nanoparticles on soil bacteria is progressively becoming apparent, yet remains inadequately comprehended. Factors such as particle size, surface charge, capping agents, the presence of divalent anions and cations, as well as the composition and charge of the bacterial cell wall, contribute to this complexity [55,57]. Previous research has indicated that the utilization of organic amendments, such as manure, organic materials, and microbial additives, can lead to a reduction in soil bacterial activity and a decrease in microbial diversity [58,59]. These declines in both abundance and diversity may be linked to the capacity of soil amendments to modify the soil pH, nutrient dynamics, and the bioavailability of heavy metals, which can favor the proliferation of certain bacterial populations while simultaneously diminishing the presence of less common soil species (see Figure 1 and Figure 3). Additionally, the reduction of rare species significantly impacts the α-diversity of rhizospheric microbes [60], and altered soil conditions may disrupt established competitive interactions among microbial communities [61]. Some bacterial taxa may exhibit greater resilience to these environmental shifts, while others may struggle to adapt, potentially conferring a competitive advantage to specific microbes and resulting in an overall reduction in diversity [59,62]. Consequently, dominant species may outcompete others for essential resources and ecological niches [63], culminating in a decrease in diversity and a transition toward a microbial community characterized by a limited number of species. Despite reduced α-diversity, enriched Firmicutes may offset functional losses by enhancing Cd immobilization. Long-term monitoring of nitrogen-cycling taxa is recommended.

Principal Coordinate Analysis (PCoA) based on the Bray–Curtis distance allowed for a clear distinction between the microbial compositions of soil samples collected from different treatments (Figure 5). The results of PCoA illustrated the differences in bacterial community structure between the flooded and drained phases, explaining a cumulative percentage variance of 102% for the species (PC1 accounts for 67%, while PC2 accounts for the remaining 35%). The community structure of the soil samples differed significantly between the flooded and drained phases. The bacterial communities in the nanomaterial-treated samples were significantly separated from those in the control group during the drainage phase. Furthermore, PCoA revealed that treatments with added nanomaterials formed a relatively compact cluster, distinct from the control group. These results indicate that the addition of nanomaterials had a minimal impact on the structure of the soil microbial community, while soil water management emerged as the primary factor driving changes in microbial community structure.

We selected microflora at the phylum taxonomic level to analyze the composition and abundance of microbial communities in this study. In total, ten major phyla with relative abundance values greater than 1% were detected in all analyzed samples. At the phylum level, Firmicutes, Acidobacteriota, and Proteobacteria were the three most abundant phyla across all samples, and the bacterial community composition was consistent across different water management and nanomaterial treatments. The combined relative abundance of Firmicutes, Acidobacteriota, and Proteobacteria across all treatments ranged from 81.9% to 89.5% (Figure 6). During the reducing phase, Firmicutes exhibited the highest relative abundance, ranging from 36.7% to 50.4%. The treatments with 0.5% NCC, 1.0% NCC, 0.5% C+P, and 1.0% C+P significantly increased the relative abundance of Firmicutes by 11.0%, 2.7%, 8.0%, and 10.3%, respectively, while the 0.5% NHAP treatment resulted in a reduction of 2.6% compared to the control group (p < 0.05). In the oxidation phase, Firmicutes again displayed the highest relative abundance, ranging from 43.0% to 58.9%. The 0.5% NCC, 1.0% NCC, 1.0% NHAP, 0.5% C+P, and 1.0% C+P treatments significantly increased the relative abundance of Firmicutes by 0.9%, 15.9%, 9.4%, 16.6%, and 14.8%, respectively, compared to the control group (p < 0.05). These results suggest that Firmicutes are highly resistant to fluctuations in soil water content, with variations in their abundance primarily attributed to the addition of nanomaterials. Previous studies have indicated that Firmicutes bacteria are involved in Cd transformation and typically exhibit a relatively higher abundance in contaminated soils [64]. Firmicutes encompass numerous populations of plant growth-promoting bacteria (PGPR) that may play a crucial role in enhancing plant growth, tolerance, and Cd enrichment. The enrichment of Firmicutes (e.g., Geobacter spp.) correlates with Fe(III) reduction and Cd co-precipitation, as evidenced by their metabolic roles in dissimilatory iron reduction [64]. Future metagenomic studies are needed to confirm these pathways. This study linked microbial community shifts (e.g., Firmicutes dominance) to Cd immobilization during drainage, a phase critical for rice grain Cd accumulation. In contrast to Firmicutes, the relative abundance of Proteobacteria and Acidobacteria showed a decreasing trend compared to the control group, with the exception of the NHAP treatments. Proteobacteria and Acidobacteria are among the bacterial groups involved in the transformation of Cd in contaminated soils [65]. They are also common microorganisms in Cd-contaminated soils and demonstrate some level of tolerance to Cd [21]. While NCC reduced Cd bioavailability, its negative impact on bacterial diversity (e.g., Acidobacteria decline) may impair soil functions like organic matter decomposition. Field trials across diverse soil types are essential to validate these findings.

3.5. Relationship Between Microbial Community Structure and Environmental Factors

During the reducing phase, the eigenvalues of axes 1 and 2 of the RDA biplot were 0.607 and 0.089, respectively, explaining 69.6% of the variation in the soil microbial community (Figure 7). Consequently, the concentrations of Fe and Mn, as well as the acid-Cd and reducible-Cd fractions, were identified as key factors influencing the composition of the microbial community during this phase. The bacterial community compositions in the 1.0% NCC, 0.5% C+P, and 1.0% C+P treatments were distinctly different from those in the control group (CK), showing a positive correlation with pH, oxidizable-Cd, and residual-Cd fractions, while exhibiting a negative correlation with the reducible-Cd fraction. Additionally, in the 1.0% NCC, 0.5% C+P, and 1.0% C+P treatments, pH was closely related to the abundance of Firmicutes and Proteobacteria, suggesting that these treatments played a critical role in influencing these parameters. The bacterial communities in the CK and 0.5% NHAP and 1.0% NHAP treatments clustered together, which was significantly different from those in soils treated with 0.5% NCC. Furthermore, the bacterial communities in the CK and 0.5% NHAP and 1.0% NHAP groups exhibited positive associations with Eh, dissolved Cd, and the Cd release rate. During the oxidation phase, the explanatory variables of axes 1 and 2 in the RDA biplot accounted for 53.7% and 10.5% of the variance, respectively. The pH, acid-Cd, oxidizable-Cd, and reducible-Cd were the primary factors influencing the composition of the microbial community during this phase. The bacterial community compositions in the treatments of 0.5% NCC, 1.0% NCC, 0.5% C+P, and 1.0% C+P were distinctly separated from the control (CK), indicating that the microbial community compositions under the 0.5% NCC, 1.0% NCC, 0.5% C+P, and 1.0% C+P treatments were similar and positively influenced by pH, acid-Cd, oxidizable-Cd, and residual-Cd. In contrast, the bacterial communities in the CK treatment and the 0.5% NHAP and 1.0% NHAP treatments clustered together, exhibiting positive associations with Eh, Fe, and Mn concentrations, dissolved Cd, Cd release rate, and reducible-Cd. Additionally, pH and residual-Cd were positively correlated with Firmicutes, while Eh, Fe, and Mn concentrations, dissolved Cd, and Cd release rate were positively associated with Acidobacteria. Soil physicochemical properties play a critical role in shaping the soil microbial communities [66]. Similarly, our results demonstrated positive correlations between the NCC and C+P treatments, pH, and Firmicutes, primarily due to the addition of nanomaterials, which altered the soil environmental variables.

4. Conclusions

During both the reduction and oxidation phases, the application of NCC demonstrated the most significant impact on increasing pH levels. The incorporation of nanomaterials can transform the mobile Cd fraction into a more stable form. The proportion of residual fractions increased with the addition of nanomaterials, particularly in the NCC treatments. Both NCC and C+P effectively reduced the release rate of Cd from paddy soils. However, the diversity of the bacterial community was significantly diminished with the application of all nanomaterials. Notably, the abundance of Firmicutes bacteria increased with the addition of NCC and C+P, which play a crucial role in facilitating the transformation of Cd fractions in the soil and reducing Cd bioavailability. Overall, NCC and C+P are effective amendments for Cd-contaminated soils and can significantly decrease Cd bioavailability in acidic soils, particularly during the oxidation phase of paddy soils. This study innovatively evaluated the efficacy of NCC and NHAP in immobilizing Cd in paddy soils under simulated flooding–drainage cycles, highlighting NCC’s significant role in reducing Cd bioavailability by modulating pH and redox dynamics, alongside enhancing the microbial community structure for mitigating Cd risks in acidic soils. Further field experiments are necessary to comprehensively evaluate the efficacy of NCC and C+P for the safe production of rice.