1 Introduction

In the agricultural environment, cadmium (Cd) is one of the most hazardous heavy metals (Xiao et al. 2023). Cd contamination in paddy soil is a growing concern because it poses a serious threat to global food security and public health (Liu et al. 2021; Wang et al. 2022). According to the national survey bulletin of soil contamination in 2014, approximately 19.4% of China’s agricultural soil contains organic contaminants or inorganic heavy metals exceeding the standard, in which the over-standard rate of soil Cd contamination was 7.0% (Huang et al. 2021). Therefore, an urgently effective strategy is required to address the risks of Cd contamination to the safety of rice.

With the advancements of nanotechnology, iron-based nanomaterials have been extensively utilized to treat Cd-contaminated paddy soil (Yang et al. 2021a). Among them, nanoscale zero-valent iron (nZVI) has garnered the most interest owing to its unique nano core–shell structure, and its application potential for efficient stabilization of heavy metals (Yang et al. 2021a). Nevertheless, there are also some inherent defects in nZVI, such as easy passivation, aggregation, and poor electron selectivity (Liu et al. 2022). To overcome these limitations, some modifications of nZVI have been proposed, including loading (Jin et al. 2023), bimetallic (Dong et al. 2022), surface coating (Zhang et al. 2022b), and sulfidation (Liang et al. 2021b). Especially, sulfidized nZVI (S-nZVI) is considered an excellent modification method and gained intensive research in recent years (Liang et al. 2021b; Yuan et al. 2022). FeSx layer generated from S-nZVI not only inhibits the oxidation of nZVI and prolongs its reactive lifetime, but also promotes the transfer of electrons from the Fe0 core to the surface layer and enhances the reaction ability of nZVI (Yuan et al. 2022). Moreover, FeSx can immobilize heavy metals by adsorption, co-precipitation, ion exchange, etc. (Gao et al. 2022b). However, S-nZVI particles still easily agglomerate due to their hydrophobicity, magnetic, and nano properties (Gao et al. 2022a). As a porous structure carbonaceous solid, biochar (BC) is commonly used as a supporting material for iron-based nanomaterials (Yang et al. 2021a). In addition, BC has been regarded as a green and effective remediation agent for soil heavy due to its unique properties, such as high surface area, abundant pore structures, terrific amount of functional groups, and huge cation exchange capacity (Ma et al. 2022a). Therefore, it is expected that loading S-nZVI on BC supported not only increase the dispersion of S-nZVI and enhance its immobilization to Cd. The application of S-nZVI/BC nanomaterials for the Cd remediation of contaminated paddy soil is still uncommon, despite some experimental researches on the reduction of heavy metal pollution or the degradation of organic pollution (Gao et al. 2022a, b).

A successful immobilization technology is closely related to the migration and transformation capacity of contaminants in environmental media (Ma et al. 2022a; Yang et al. 2021a). The more active heavy metals speciation, such as the acid soluble speciation, which exhibited high mobility and bioavailability, was typically more likely to be altered in the environment. In contrast, the stability phase (residual speciation) was less mobile and higher safety, with a lower environmental risk (Jin et al. 2023). Hence, the modified Community Bureau of Reference (BCR) sequential extraction method was adopted to examine variations in Cd chemical speciation. Besides, the metals availability and leaching toxicity to be tested in the soil can be considered as comprehensive indicators to evaluate the biological effects of heavy metals during the remediation process, which has significant scientific value for comprehending how S-nZVI/BC reduces the potential ecological risks of heavy metal during the remediation of contaminated soil (Jiang et al. 2022a). Furthermore, the surface charge of soil colloids or nanomaterials can change depending on the pH of the soil, which is a critical element in controlling the tendency of heavy metals to migrate (Liu et al. 2022; Wang et al. 2021b). Notably, soil redox potential (Eh) was important to the mobility of heavy metals, the decrease of Eh was contributed to the generation of insoluble heavy metal compounds (Yang et al. 2021a). Also, previous researches have found that soil organic matter (OM) may influence the chemical behaviors of heavy metals such as sorption–resolution, complexation, and solution–precipitation (Fan et al. 2023). However, valid information about S-nZVI/BC caused alterations in soil pH and OM and the influence of pH and OM changes on Cd immobilization are still unclear, and further researches are necessary.

The research goals are listed as follows: (i) synthesize and characterize S-nZVI/BC nanomaterials; (ii) explore the effect of S-nZVI/BC on Cd immobilization by detecting the change of Cd speciation in contaminated paddy soils; (iii) evaluate the ecological risk of S-nZVI/BC by measuring the availability and leaching toxicity of Cd; (iv) analyze the possible interaction mechanisms between the S-nZVI/BC and Cd by further determining soil pH, Eh, OM, and S-nZVI/BC. The theoretical underpinnings for the efficient application of S-nZVI/BC in the treatment of Cd-contaminated paddy soil are provided by this work.

2 Materials and methods

2.1 Chemical reagents and soil properties

The wheat straw was purchased from Yangzhou YangTai Medical Instrument Co., Ltd (Jiangsu, China). Iron trichloride hexahydrate (FeCl3·6H2O, 99.0%), sodium borohydride (NaBH4, 98.0%), sodium dithionite (Na2S2O4, 99.0%), hydrogen nitrate (HNO3), and absolute alcohol (C2H5OH) were obtained from Sinopharm Chemical Reagent Co., Ltd (Shanghai, China). Chemicals of analytical grade were employed without additional purification throughout the experiment. To prepare all solutions, ultrapure water (18.2 MΩ·cm) was utilized. Trial soil was provided from the surface soil (0–20 cm) of a paddy field in Hanjiang, Yangzhou, Jiangsu Province, China. Supporting Information (SI) provided specifics regarding the soil collection and subsequent treatment.

2.2 Synthesis and characterization of nanomaterials

BC was prepared according to the method provided by modified Gao et al. (2022a). Specifically, BC was derived pyrolysis of crushed wheat straw placed in a quartz boat in a tubular furnace with a nitrogen-filled environment. The pyrolysis process was elevated to 600 °C at 5 °C·min−1 and then maintained the temperature under nitrogen protection for 2 h. Finally, ground the solid into powder after cooling to room temperature, and collected after passing through a 0.15-mm sieve to obtain the desired BC.

S-nZVI/BC was synthesized by using NaBH4 as the reducing agent to reduce Fe3+ solution (Jiang et al. 2022b). Firstly, after transferring 100 mL of 2.4152 g FeCl3·6H2O solution to a three-necked flask and adding 1.0 g of BC to the flask, the mixed solution was stirred for 30 min under a flow of nitrogen to remove the residual air in the solution. Secondly, 1.6 g of NaBH4 and 0.1942 g of Na2S2O4 were prepared as a 100-mL mixed solution, and the solution was added dropwise to the continuously stirred solution in the three-necked flask, then after dropping, continue stirring for 30 min to full reaction. Subsequently, a magnet was used to separate solids and liquids in the reaction solution, then discarded the supernatant, washed the remaining black precipitates with ethanol three times to remove impurities, and the black precipitates were dried in a vacuum drying box for 8 h (60 °C) to obtain the S-nZVI/BC. Finally, S-nZVI/BC was ground and stored for later use. The preparation flow chart of the S-nZVI/BC was depicted in Fig. 1. The preparation procedure of nZVI, nZVI/BC, and S-nZVI were similar to that described above, nZVI was synthesized without adding BC and Na2S2O4, nZVI/BC was attained without adding Na2S2O4, and S-nZVI was obtained without adding BC. The pH of the nanomaterials and the characterization details of the nanomaterials before and after the reaction were shown in the SI.

Fig. 1
figure 1

Flow chart for the preparation of S-nZVI/BC nanomaterial

Full size image

2.3 Experiment design

For paddy soil remediation, the appropriate amount of stored dry paddy soil was placed in a 50-mL-circled plastic centrifuge tube, then the nanomaterials were incorporated into the soil at a ratio of 0.03:1 mass of nanomaterials to soil, and all the samples were shaken uniformly and mix thoroughly. Then, an appropriate amount of ultrapure water was added to simulate the state of a paddy field during flooding, and the soil-to-water ratio was 1:5. The control group experiment was conducted without any nanomaterial and the circumstances for the experiments were unchanged. The soil samples were assayed at 1, 7, 14, 21, 35, and 49 days in order to specifically examine how reaction time affects Cd immobilization. During the treatment period, each treatment group was cultured at room temperature, and its water holding capacity was continuously maintained by adding ultrapure water. After reaching the set treatment time, the paddy soil samples were centrifuged for further analysis.

2.4 Determination of Cd speciations, availability, and leachability

The Cd speciation in contaminated soil was identified by the BCR continuous extraction method proposed by Liu et al. (2018). Acid soluble, reducible, oxidizable, and residual are the four speciations of Cd that exist in soil. Usually, 0.11 M CH3COOH and 0.5 M NH2OH·HCl were used to extract the acid soluble and reducible, respectively. The oxidizable was extracted using H2O2 and 1 M NH4OAc. The residual was digested with HNO3-HF-HClO4 (5: 4: 2) system. The soil Cd availability was assessed using the DTPA extraction method (Yang et al. 2021c). The TCLP test assessed Cd leachability (Liu et al. 2018). More details about the extraction steps were displayed in SI and Table S1.

2.5 pH, Eh, and OM measurement

A pH meter was employed to measure soil pH at the soil-to-water ratio of 1:5 (Puissant et al. 2019). The soil Eh value was measured and recorded by inserting compounding electrode in the soil. To clarify the contents of soil OM, the K2Cr2O7 volumetric method was employed and the content of OM was obtained based on (Wang et al. 2021b):

$$\begin{array}{c}{\text{OM}}=\left[\frac{0.5\times ({V}_{0-}V)\times 0.001\times 3.0\times 1.33}{m}\right]\times 1000\times 1.725\end{array}$$
(1)

where V0 is the FeSO4 consumption of the blank, V is the FeSO4 consumption of the sample, and m is the weight of the soil.

2.6 Date process and statistical analyses

All experiments in this study were conducted three times and all outcomes were expressed as mean values ± standard deviation and drawn with Origin 2018 software (Northampton, MA, USA). Version 26.0 of the IBM SPSS Statistics for Windows (Armonk, NY, USA) was used for one-way analysis of variance (ANOVA) and Pearson correlation analysis.

3 Results and discussion

3.1 Characterization

The morphologies of nZVI, BC, S-nZVI, S-nZVI/BC by the scanning electron microscope (SEM) images were displayed in Fig. 2. The pristine nZVI particles were spherical and agglomerated together due to magnetic effect (Fig. 2a). From Fig. 2b, BC contained abundant pore structures, which could effectively disperse S-nZVI and provide more adsorption sites for heavy metals. S-nZVI particles displayed a rougher surface than nZVI, owing to the formation of an outer layer that contains a large amount of iron sulfide and a small amount of iron oxide (Fig. 2c) (Bhattacharjee and Ghoshal 2018; Han et al. 2019). Also, according to Fig. 2d, S-nZVI particles adhered to BC surfaces and displayed effective dispersion. The transmission electron microscopy (TEM) images suggested that the nZVI particles presented chain-like structure with size of about 50 nm (Fig. 2e). The TEM images of S-nZVI/BC revealed a sphere particle phase, where particles were remarkably dispersed (Fig. 2f). Significantly, BC was observed in the vicinity of the sphere particles.

Fig. 2
figure 2

SEM image of nZVI (a), BC (b), S-nZVI (c), S-nZVI/BC (d), and TEM image of nZVI (e), S-nZVI/BC (f)

Full size image

The X-ray photoelectron spectroscopy (XPS) spectrum of S-nZVI/BC was presented in Fig. 3. During the survey spectrum analysis (Fig. 3a), the coexistence of Fe, C, O, and S elements was observed in S-nZVI/BC. Fe 2p3/2 and Fe 2p1/2 were visible at Fe 2p spectrum peaks in Fig. 3b, where the characteristic peaks at 708.9 eV and 719.7 eV were considered as the Fe0 (Wang et al. 2021a), 710.1 eV and 724.5 eV were represented to the Fe2+ (Wang et al. 2021a), 713.0 eV and 731.2 eV were indicated the peaks of Fe3+ (Rao et al. 2021). The peak areas of 24.0%, 49.4%, and 26.6% for Fe0, Fe2+, and Fe3+, respectively, demonstrate that Fe2+ was the dominant iron species in the nanomaterial. Peaks situated at 169.2, 167.8, 163.9, 162.9, 162.4, and 160.9 eV were ascribed to the S compounds (SO42−, SO32−, Sn2−, S22− and S2−) with corresponding peak areas of 14.9%, 22.2%, 15.5%, 22.6%, and 24.8%, respectively (Fig. 3c) (Luo et al. 2022; Singh et al. 2021). The creation of a FeSx-dominated layer was indicated by the detection of S2−, S22−, and Sn2− species, which may have increased the reactivity of nanomaterial (Gao et al. 2022b). Also, the presence of SO32− and SO42− species suggested that some of the S were oxidized to form sulfur oxides during the synthesis process (Gao et al. 2022a). As depicted in Fig. 3d, it was determined that − COOH, C = O, C–O, and C = C were responsible for peaks at 293.5, 288.8, 284.4, and 283.9 eV of C1s (Fan et al. 2021; Wang et al. 2021a), respectively. Figure 3e displayed the peaks of O 1s at 532.2, 531.1, and 530.4 eV, considered as C = O, C–O, and -OH of BC (Khan et al. 2021; Luo et al. 2019), respectively. While the peak at 531.8 and 529.5 eV were corresponded to the presence of FeOOH and Fe–O in the nanomaterial (Hu et al. 2021; Zhu et al. 2020).

Fig. 3
figure 3

XPS spectra of S-nZVI/BC: survey (a), Fe 2p (b), S 2p (c), C 1s (d), O1s (e), and FTIR spectra of S-nZVI/BC (f)

Full size image

The Fourier transform infrared spectroscopy (FTIR) spectra of nZVI, BC, S-nZVI, nZVI/BC, and S-nZVI/BC were shown in Fig. 3f. The − OH bending vibration at 1404 cm−1 was observed in nZVI, which might be the hydroxylation reaction on nZVI surface (Tang et al. 2021). The bands at about 3500 cm−1 and 3700 cm−1 were explained by fluctuations in the O–H bonding of adsorbed water or the hydroxyl groups (− OH) stretching on the BC surface (Tang et al. 2021). In addition, the C = O stretch at 1667 cm−1 (Luo et al. 2019), the − CH2 stretch at 1470 cm−1 (Huang et al. 2020) and the − OH bending vibration at 1265 cm−1 were assigned to BC (Liu et al. 2018). The stretching vibrations of C–O and C–H were displayed at 910 cm−1 and 829 cm−1, respectively, demonstrating the formation of organic compounds such as phenols and alcohols during the pyrolysis of wheat straw (Li et al. 2020; Luo et al. 2019). The diffraction peak of FeS2 appeared at 1132 cm−1 in S-nZVI/BC, indicating that the sulfidation was successful (Tang et al. 2021). The C-O (910 cm−1) peak of BC was found to be weakened and the C-H peak was shifted from 829 to 814 cm−1 in the spectra of nZVI/BC and S-nZVI/BC, suggesting the binding of Fe0 with some functional groups during the synthesis of composites (Luo et al. 2019). The FTIR analysis obtained coincides with the experimental data of SEM, TEM, and XPS. The above characterization results demonstrate the successful synthesis of S-nZVI/BC in this work.

According to the results of the elemental analysis (Table S2), S-nZVI and S-nZVI/BC materials had successfully been sulfidized as evidenced by the much higher S element content in comparison to nZVI, BC, and nZVI/BC. However, the S content of S-nZVI/BC decreased after BC supporting, but the C/H atomic ratio of nZVI/BC and S-nZVI/BC increased, demonstrating the efficacy of the supporting of nZVI and S-nZVI onto the BC surface. Moreover, to enhance understanding of nZVI, BC, nZVI/BC, S-nZVI, and S-nZVI/BC, their pH was measured before batch experiments (Table S3), with values of 8.60, 10.39, 8.91, 8.75, and 9.29, respectively.

3.2 Changes in the behavior of Cd

According to the BCR sequential extraction procedure, the stability of Cd speciation followed: residual > oxidizable > reducible > acid soluble (Rao et al. 2021). It was also observed that Cd speciation changed in soil during the treatment period, as shown in Fig. 4a. In the control group, the majority of the Cd speciation was acid soluble speciation which amounts to approximately 65.5%, while the percentage of reducible, oxidizable, and residual speciation were about 30.90%, 2.38%, and 1.22%, respectively. It indicates that Cd in the original soil has high mobility and high environmental risk. The speciation distribution of Cd in each treatment group presented a similar change trend during the whole treatment period. From Fig. 4a, the oxidizable and residual speciation of Cd were time-dependent increased while the acid soluble speciation of Cd was time-dependent decreased, compared to the irregular change of Cd reducible speciation with time. As a result, the presence of nZVI, BC, nZVI/BC, S-nZVI, and S-nZVI/BC treatments significantly reduced by 19.77%, 15.92%, 26.95%, 25.52%, and 36.05% of the acid soluble of Cd after 49 days, respectively. On the contrary, the percentages of oxidizable speciation of Cd were elevated from 2.38% to 8.34%, 6.23%, 6.23%, 6.29%, and 6.16% and residual speciation increased from 1.22% to 13.68%, 11.84%, 21.32%, 18.83%, and 32.04% with the treatment of nZVI, BC, nZVI/BC, S-nZVI, and S-nZVI/BC, respectively. There was relatively less change in the percentages of reducible and oxidizable speciation compared to acid soluble and residual speciation, indicating that the acid soluble speciation was mainly transformed into the residual speciation after nanomaterials remediation. From Fig. 4b and c, the acid soluble speciation of Cd treated with nZVI and S-nZVI/BC was 0.27 and 0.53 times lower than the control group. The other Cd speciations (reducible, oxidizable, residual) also increased in the nZVI group compared to the control group by 0.05, 1.66, and 3.05 times (Fig. 4b), while the treatment of S-nZVI/BC was 0.05, 0.89, and 8.51 times (Fig. 4c), respectively. Thus, the addition of these nanomaterials effectively reduced mobility of Cd in soil, particularly when combined with S-nZVI/BC. At present, some amendments applications for the Cd remediation on soil contaminated were also reported by others, which was summarized in Table S4. For example, Yang et al. (2022) confirmed that compost palygorskite composite treatment increased the residual speciation of Cd by 7.74%. Song et al. (2022) investigated the soil Cd immobilization by biochar-supported nanoscale zero-valent iron (nZVI@BC) composite. Results indicated that the treatment of contaminated soil samples with nZVI@BC at a dosage of 3% (w/w) for 40 days increased Cd residual speciation by 9.26%. Ma et al. (2022b) demonstrated that the modified zeolite (MZEO) facilitated Cd immobilization in Baiyin soil, with a 15% increase in Cd residual specifications during 56 days of treatment. Compared to these amendments, S-nZVI/BC exhibited better Cd remediation effect, and its application could effectively reduce the mobility of Cd.

Fig. 4
figure 4

a Change in the speciation of Cd during 49 days of treatment. (From left to right is control group, nZVI, BC, nZVI/BC, S-nZVI, S-nZVI/BC). The percentage of different speciation of Cd decreased or increased times with the treatment of nZVI (b) and S-nZVI/BC (c) at 49 days compared to control group. Error bars medicate standard deviation (n = 3)

Full size image

DTPA extraction was developed to predict heavy metal availability in soil. Figure 5a described the transformation of the DTPA-extractable Cd contents in the soil after treatment with different nanomaterials. The average amount of DTPA-extractable Cd in the control group essentially remained at 2.14 mg·kg−1 across the treatment process, with an initial content of DTPA-extractable Cd was 2.13 mg·kg−1 and a content of 2.16 mg·kg−1 after 49 days. Treatment groups had significantly lower levels of DTPA-extractable Cd than control groups (P < 0.05) and their levels decreased with each treatment day. From Fig. 5a, compared with control group, the DTPA-extractable Cd content were decreased by 1.28, 0.98, 1.47, 1.45, and 1.70 mg·kg−1 for nZVI, BC, nZVI/BC, S-nZVI, and S-nZVI/BC, respectively with the immobilization efficiencies of 45.62%, 59.63%, 67.80%, 68.71%, and 79.59% after 49 days of treatment, respectively (Fig. 5b). The result indicated that the addition of the nanomaterials diminished the content of DTPA-extractable Cd. Among them, S-nZVI/BC had the lowest DTPA-extractable Cd content and highest immobilization efficiencies, confirming that S-nZVI/BC displayed better immobilization potential to decrease the availability of Cd. Furthermore, the DTPA-extractable Cd content and percentages of acid soluble speciation exhibit basically the same trend of being lowered, going against the increasing tendency for the percentages of residual speciation. This situation concurred with the description of Fuentes et al. (2004), who reported that the acid soluble speciation obtained by BCR extraction was compared with the Cd content obtained by DTPA extraction, which could be pointed out that the change level of the metals was similar. Fan et al. (2020) also highlighted that the thiol-modified rice straw biochar increased Cd retention in soil complexing with mobilized Cd speciations, decreasing DTPA-extractable Cd levels. As a result, the lower amount of DTPA-extractable Cd showed that the activated Cd was transformed into a more stable species.

Fig. 5
figure 5

Changes in content (a) and the immobilization efficiencies (b) of DTPA-extractable Cd, the concentration (c), and the immobilization efficiency (d) of TCLP-leachable Cd. Different lowercase letters represent significant differences within each treatment using one-way ANOVA (p < 0.05)

Full size image

TCLP is usually employed to evaluate the performance of heavy metal leaching and migration in solid media. The variation of the TCLP-leachable Cd concentration and the immobilization efficiency in the soil treated with nanomaterials were exhibited in Fig. 5c and d. In the control group, it was found that the Cd concentration was slightly decreased, the initial TCLP-leachable Cd concentration was decreased from 0.079 to 0.075 mg·L−1 after 49 days of treatment. The concentration of TCLP-leachable Cd in soil was obviously lower after adding nanomaterials than the control group. From Fig. 5c, the concentration of TCLP-leachable Cd decreased to 0.059, 0.063, 0.051, 0.052, and 0.035 mg·L−1 for nZVI, BC, nZVI/BC, S-nZVI, and S-nZVI/BC, respectively, corresponding to 25.32%, 20.25%, 35.44%, 34.18%, and 53.7% immobilization efficiencies (P < 0.05) (Fig. 5d), indicating that nZVI, BC, nZVI/BC, S-nZVI, and S-nZVI/BC may be able to reduce the biological toxicity of Cd in soil. S-nZVI/BC exhibited a superior immobilization capacity of TCLP-leachable Cd than nZVI, BC, nZVI/BC, and S-nZVI. The above analysis and discussions suggested that nanomaterials could contribute to effectively decreasing TCLP-extractable Cd concentration, reducing the dissolution of Cd in soils, and thereby reducing the environmental risk.

This study also synthesized four micromaterials, i.e., micro zero-valent iron (mZVI), BC supported mZVI (mZVI/BC), sulfurized mZVI (S-mZVI), and BC supported S-mZVI (S-mZVI/BC), for the immobilization of Cd in paddy soil. The characterization results of the micromaterials and the immobilization effect on Cd were shown in Figs. S1 and S2, respectively. Results indicated that micromaterials could reduce the content of DTPA-extractable Cd and the leaching toxicity of Cd, but were not as effective as nanomaterials.

3.3 Changes of pH, Eh, and OM in paddy soil

It is widely accepted that soil pH has a significant effect on the mobility of heavy metals (Xu et al. 2021). Usually, lower pH decreases the Cd adsorption on soil colloids by reducing the negative charge carried by soil colloids, which enhances the mobility of Cd (Liu et al. 2022). Hence, Fig. 6a represented the change in soil pH during the course of the 49 days treatment. The pH of the control group remained unchanged basically in the treatment process, which was approximately 7.57 ± 0.03. Compared to control group, the application of nZVI, BC, nZVI/BC, S-nZVI, and S-nZVI/BC considerably higher soil pH on the first day, corresponding to increase from 7.57 to 8.50, 7.82, 8.10, 8.52, and 8.36, respectively (P < 0.05). The alkaline characteristics of BC and the production of OH− by the oxidation process of Fe0 were the main increased causes of pH (Gong et al. 2021). In addition, it was observed that the change of pH in other treatment groups during 49 days was irregular. It has been reported that OH− could combine with dissolved metal ions to release H+, reducing the soil pH (Li et al. 2019). Moreover, the pH decreases in nZVI and modified nZVI treatment groups may be attributed to the hydrolysis of dissolved iron ions (Li et al. 2019). Specifically, Cd and HS− formed CdS also released H+ in the S-nZVI and S-nZVI/BC treatment groups (Liang et al. 2022). However, compared to the blank group, the pH of the treated group was higher during the whole treatment process. After 49 days, the pH values of nZVI, BC, nZVI/BC, S-nZVI, and S-nZVI/BC treatment groups were 8.35, 8.25, 8.28, 8.46, and 8.26, and increased by 0.76, 0.66, 0.69, 0.87, and 0.67 units compared with control group, respectively. The pH of each treatment group was still high than 8 after 49 days of treatment. It has been revealed that the formation of Cd hydroxyl substances occurred at pH ≥ 8 (Xue et al. 2018). On the other hand, the negative charge of the nanomaterials increased at high pH, enhancing the electrostatic affinity to positively charged metal ions and strengthening the adsorption and complexation, which facilitated the immobilization of Cd (Yang et al. 2021c).

Fig. 6
figure 6

Changes of pH (a), Eh (b), and OM (c) during 49 days of treatment in amendment soil. Different lowercase letters represent significant differences within each treatment using one-way ANOVA (p < 0.05)

Full size image

Changes of soil Eh will lead to changes in soil redox conditions, which will further influence the transformation of heavy metals (Yao et al. 2022). It was observed from Fig. 6b that the Eh of the control group increased firstly and then decreased during the treatment period. The Eh value increased by only 36 mV after 49 days of treatment compared to the Eh value on the first days. The reason for this phenomenon might be that, when soil samples were submerged at the early stage, there were microorganisms that could serve as a medium for oxygen transfer from the water, thereby contributing to the maintenance of soil oxidation state. However, with the intensification of aerobic microbial activity in the soil, oxygen is progressively depleted, causing a corresponding decrease in Eh values (He et al. 2022). After adding nZVI, BC, nZVI/BC, S-nZVI, and S-nZVI/BC to the contaminated soil samples, the Eh values of the nZVI, nZVI/BC, and S-nZVI/BC treatment groups were in a strong reducing state on the first days, at − 652, − 626, and − 654 mV, respectively. Then, they gradually increased over time, and the values were − 216, 83, and 169 mV after 49 days, respectively. The Eh values of the BC and S-nZVI treatment groups gradually decreased in the early stage but began to rise after 14 days. The changes of Eh were not only related to the effects of microbial transport and oxygen consumption mentioned earlier but also speculated to be related to the release and hydrolysis of Fe2+ (Crane and Sapsford 2018). When the soil was in a reducing state, it became easy for Cd to combine with sulfides to form insoluble CdS, and the adsorption/co-precipitation effect of Fe–Mn (oxyhydro)oxides on Cd was enhanced, ultimately reducing its mobility. Furthermore, secondary Fe minerals formed by the anaerobic oxidation of Fe(II) could effectively adsorb free Cd (Wang et al. 2019). Over time, Fe2+ was oxidized to form Fe3+, and the Eh of each treatment group increased (Shen et al. 2020). With the increase of Eh value, sulfide-bound Cd may be released, and the released Cd2+ can then be further stabilized by the nanomaterial (Chen et al. 2020). The soil gradually transitioned to an oxidized state, but the Eh values of each treatment group were lower than those of the control group throughout the treatment period.

As shown in Fig. 6b, the OM content of the control group was lower than that of each treatment group during the first day. As already mentioned, BC was prepared by high-temperature pyrolysis of wheat straw in this experiment. It not only contained a high carbon content but also exhibited strong adsorption potential for organic molecules, increasing the content of soil OM (Fan et al. 2020; Zhang et al. 2022a). For nZVI, it released iron ions which may be combined with water-soluble substances in the solid phase to enhance the soil OM content (Hui et al. 2021). Furthermore, the sulfidation reduced the release of Fe2+ in S-nZVI and most of the Fe2+ ions were entrapped in the outer layer of the particles (Cheng et al. 2019), so the OM content of S-nZVI treatment group only increased by 0.23 mg·kg−1 compared with the control group. After that, the OM content in each treatment group showed a downward trend and become stable. The OM content was 21.15, 23.80, 21.16, 18.76, and 22.76 mg·kg−1, respectively, for nZVI, BC, nZVI/BC, S-nZVI, and S-nZVI/BC, which were higher than control group (20.30 mg·kg−1) except for S-nZVI. This might be attributed to the increased OM content promoting microbial activities in soil (Wang et al. 2021b). Moreover, the changes in Cd speciations indicated that the nanomaterials could gradually facilitate the transfer of mobility speciation to stable speciations and decrease Cd toxicity (Kong and Lu 2022), altering soil microbial community structures and functions, promoting the activity of some microorganisms that decompose OM, thus reducing the content of OM (Xu et al. 2021).

To further explore the relationship between soil properties and heavy metal mobility, Pearson correlation analysis among the Cd speciation, the content of extracted Cd, and pH, Eh, and OM were shown in Table 1. The pH showed significant negative correlation with the Cd acid soluble speciation and the DTPA-extractable Cd, while a significant positive correlation with the Cd oxidation speciation, with correlation coefficients of − 0.823 (P < 0.05), − 0.930 (P < 0.01), and 0.836 (P < 0.05), suggesting that the increase of pH could decrease the mobility of Cd, and then reduced the environmental risk of Cd. There was a negative correlation between the Eh and the Cd oxidation speciation, with a correlation coefficient of − 0.919 (P < 0.01). The soil Eh value of each treatment group was lower than that of the control group during the 49 days of treatment, corresponding to the increase of the Cd oxidation speciation in the previous BCR sequential extraction, which proved that the decrease of Eh value could promote the Cd mobility to stable state. OM could directly adsorb and complex with Cd, or indirectly affect the immobilization of Cd by promoting the dissolution of iron oxides and co-precipitating/complexation with iron oxides to produce organic matter-iron oxide complexes (Bao et al. 2022). However, no significant relationships were observed between soil OM and Cd in each speciation, or even the extracted Cd concentrations, which indicated that variation in soil OM had limited effect on Cd transformation with nanomaterial treatment. DTPA-extractable Cd had a considerably negative correlation with pH (P < 0.01). These indicated that an increase in pH facilitated the transition of Cd to the residual speciation, which reduced the toxicity and availability of Cd. In short, soil pH was the major factor of Cd mobility in soil.

Table 1 Pearson correlation analysis among the Cd speciation, the content of extracted Cd, and soil pH, Eh and OM
Full size table

3.4 Characterization of nanomaterials after remediation

Investigations into the magnetically separated particles isolated from the treated paddy soil after the reaction were studied to further explore the possible stabilization mechanism of Cd by S-nZVI/BC in soil. Those investigations included SEM, FTIR, and X-ray diffraction (XRD). Figure 7 presented the SEM images and corresponding energy dispersive spectrometry (EDS) mapping of S-nZVI/BC particles separated after reacting with Cd for 49 days. Compared with the original S-nZVI/BC (Fig. 2d), the S-nZVI/BC surface became rougher after the reaction, most of the particles were mainly comprised flaky and a litter rough spherical structure (Fig. 7a). The EDS mapping of C, O, S, Fe, and Cd was displayed in Fig. 7b−f. The distribution of Cd was clearly matched with Fe, O, and S, which illustrated that Fe, O, and S were the key elements for Cd adsorption. Furthermore, the EDS analysis revealed that the particles were mainly composed of C (12.36%), O (55.28%), S (0.08%), Fe (23.29%), and Cd (8.99%) (Fig. 7g), indicating that Cd was successfully enriched by S-nZVI/BC. Thus, it was conjectured that the soluble Cd was immobilized in the paddy soil by reacting with FeSx on the S-nZVI/BC surface to form CdS with low solubility, and/or the formation of complexes with iron oxide/hydroxide (Liang et al. 2021a).

Fig. 7
figure 7

The SEM images (a) of S-nZVI/BC after reaction with Cd, and the corresponding mapping of C (b), O (c), S (d), Fe (e), Cd (f), and EDS (g)

Full size image

S-nZVI/BC before and after the reaction was further analyzed using FTIR spectroscopy to reveal the changes in functional groups (Fig. 8a). All the bands were shown at 3700–3400 cm−1 as the result of O–H stretching vibration (Tang et al. 2021), the O–H peak after the remediation of Cd was sharper at 3632 cm−1 and a new O–H peak appeared at 3440 cm−1 (Dong et al. 2017; Tang et al. 2021), probably due to the oxidation and hydrolysis of Fe0 to produce more iron hydroxide (e.g., FeOOH) (Hui et al. 2022), as well as the combination of Cd2+ with hydroxyl groups to form various hydroxyl compounds (e.g., Cd(OH)2) (Liang et al. 2022). The C = O vibration peak of the functional group from 1667 to 1652 cm−1, presumably resulting in the changes caused by the complexation of Cd with functional groups in BC structures during soil remediation. The Cd–O vibration peak was located at 1440 cm−1, confirming the stabilization of Cd2+ on S-nZVI/BC by adsorption and/or complexation (He et al. 2022). Additionally, the characteristic peak at 1039 cm−1 was observed after remediation, probably corresponding to the stretching vibration of hydrated ferric sulfate, indicating the possible existence of sulfate ions in the S-nZVI/BC (Hui et al. 2022). Moreover, the Fe–O vibration peaks were observed after remediation at 698, 534, and 470 cm−1, inferring that more iron oxides were formed due to the oxidation of Fe0 (Gu et al. 2021; Liang et al. 2021a).

Fig. 8
figure 8

FTIR spectra (a) and XRD (b) of S-nZVI/BC before and after reaction with Cd

Full size image

The XRD patterns were conducted to determine the changes in chemical transformation of S-nZVI/BC before and after remediation of Cd-contaminated paddy soil. According to Fig. 8b, the XRD analysis of S-nZVI/BC before remediation showed an apparent characteristic peak of Fe0 (2θ = 44.9°, 65.0°, 82.1°) (Gao et al. 2022a; Yang et al. 2021b), and iron oxide/hydroxide peaks at 2θ = 21.1° (FeOOH) and 2θ = 50.3° (Fe2O3) (Liang et al. 2022; Yang et al. 2021b). It was inevitable that part of Fe0 in the nanomaterials was oxidized. The amorphous carbon of BC was observed at 2θ = 26.7°, which indicated that BC successfully supports S-nZVI (Hamid et al. 2022). The peaks associated with sulfur-containing compounds (e.g., FeS or FeS2) were not observed, probably due to their low crystallinity (Gao et al. 2020). Compared with the pristine S-nZVI/BC, the XRD patterns of S-nZVI/BC after treatment observed the intensity of the special peak at 2θ = 44.9° for Fe0 was weakened and 2θ = 65.0°, 82.1° of Fe0 almost disappeared. In contrast, the distinct peaks located at 2θ = 21.1° for FeOOH, and 2θ = 50.3° for Fe2O3 were enhanced, while some diffraction peaks at 2θ = 54.8° for FeO and 2θ = 63.9° for Fe2O3 were discovered (He et al. 2022). These iron oxide/hydroxide are beneficial to provide more reaction sites for immobilizing free Cd in soil via adsorption, complexation, and other mechanisms (Dong et al. 2017). Moreover, some novel characteristic peaks of Cd(OH)2 (2θ = 36.7°, 42.6°) and CdFe2O4 (2θ = 39.2°, 60.0°) were found, which confirmed precipitation and complexation were detected for Cd stabilization (Ainiwaer et al. 2022; He et al. 2022; Yang et al. 2021b).

3.5 Immobilization mechanism

Based on the characterization of the separated S-nZVI/BC after remediation, it was determined that the remediation mechanism of Cd included adsorption, complexation, and precipitation processes (Fig. 9). Firstly, alkaline biochar had more negative charges on its surface, which increased the electrostatic interaction between the nanomaterials and Cd (Liang et al. 2022). The abundant functional groups on the S-nZVI/BC influenced Cd adsorption through promote the formation of Cd compounds (Eqs. (2–3)) (Liang et al. 2022). Secondly, Fe0 in S-nZVI/BC was easily converted into various iron oxide/hydroxide in soil under suitable soil conditions. According to the XRD results, the oxidation product types of Fe were mainly FeOOH and Fe2O3, which possessed high adsorption capacity (Eqs. (4–6)) (Hui et al. 2022). It provided the adsorption sites for the surface complexation of iron oxide/hydroxide with Cd (FeOCd+, FeOCdOH) (Eqs. (7–8)), which was confirmed by some researchers (Liang et al. 2022). Then, Cd could replace Fe2+ from FeS to form CdS precipitation with lower solubility (Eq. (9)) (Liang et al. 2020). Also, Cd could react with the dissolved S ions to form CdS (Eqs. (10–11)) (Gao et al. 2022b). Furthermore, it was also possible that the surface complexation of FeSx layer reaction with Cd (FeS-Cd2+) (Eq. (12)) (Liang et al. 2022). In addition, the incorporation of nanomaterials raises soil pH and promotes the formation of insoluble Cd(OH)2 from Cd2+ and OH− (Yang et al. 2021b). Furthermore, the generation secondary iron mineral (CdFe2O4) had strong complexing ability to Cd (He et al. 2022).

Fig. 9
figure 9

A schematic of the possible reaction mechanism of Cd by S-nZVI/BC

Full size image
$$\begin{array}{c}\mathrm C-\mathrm{OH}+\mathrm{Cd}^2\rightarrow\mathrm C-\mathrm{OCd}^++\mathrm H^+\end{array}$$
(2)
$$\begin{array}{c}2\mathrm C-\mathrm{COOH}+\mathrm{Cd}^{2+}\rightarrow(\mathrm C-\mathrm{COO})_2\mathrm{Cd}+2\mathrm H^+\end{array}$$
(3)
$$\begin{array}{c}\mathrm{Fe}^0+2{\mathrm H}_2\mathrm O\rightarrow\mathrm{Fe}^{2+}+{\mathrm H}_2+2\mathrm{OH}^-\end{array}$$
(4)
$$\begin{array}{c}4\mathrm{Fe}^{2+}+{\mathrm O}_2+2{\mathrm H}_2\mathrm O\rightarrow4\mathrm{Fe}^{3+}+4\mathrm{OH}^-\end{array}$$
(5)
$$\begin{array}{c}\,\,4\text{F}\mathrm e^0+{\mathrm O}_2+2{\mathrm H}_2\mathrm O\rightarrow4\mathrm{FeOOH}\end{array}$$
(6)
$$\begin{array}{c}\mathrm{FeO}^-+\mathrm{Cd}^{2+}\rightarrow\mathrm{FeOCd}^+\end{array}$$
(7)
$$\begin{array}{c}\mathrm{FeOH}+\mathrm{Cd}^2+{\mathrm H}_2\mathrm O\rightarrow\mathrm{FeOCdOH}+2\mathrm H^+\end{array}$$
(8)
$$\begin{array}{c}\,\,\,\text{FeS+C}\mathrm d^{2+}\rightarrow\mathrm{Fe}^{2+}+\mathrm{CdS}\end{array}$$
(9)
$$\begin{array}{c}\mathrm{FeS}+\mathrm H^+\rightarrow\mathrm{Fe}^{2+}+\mathrm{HS}^-\end{array}$$
(10)
$$\begin{array}{c}\,\,\,\text{H}\mathrm S^-+\mathrm{Cd}^{2+}\rightarrow\mathrm H^+\mathrm{CdS}\end{array}$$
(11)
$$\begin{array}{c}\mathrm{FeSH}^-+\mathrm{Cd}^{2+}\rightarrow\mathrm{FeS}-\mathrm{Cd}^++\mathrm H^+\end{array}$$
(12)

4 Conclusion

S-nZVI/BC composites were synthesized and employed in this investigation for Cd immobilization in contaminated paddy soil. The synthesized products were characterized by SEM, TEM, XPS, and FTIR. These results confirmed that S-nZVI/BC had excellent dispersity, and the S-nZVI particles showed an obvious shell-core structure with Fe0 as the core and FeSx, sulfur oxides, and a small amount of iron oxide/hydroxide in the shell. Remediation tests in the paddy soil determined that S-nZVI/BC could effectively convert the acid soluble speciation of Cd to residual speciation. At the same time, DTPA-extractable Cd availability and TCLP-leachable Cd leachability were significantly decreased. Moreover, it was verified that the increase in soil pH caused by the addition of S-nZVI/BC could decrease the mobility of Cd. The immobilization mechanism of Cd clarified that surface adsorption or complexation on the formed iron oxide/hydroxide or FeSx layer and secondary iron mineral, precipitation to form CdS and Cd(OH)2. Furthermore, this study estimated the material cost of adding 3% S-nZVI/BC to 1 g of soil to be 0.207 RMB. Therefore, using S-nZVI/BC to remediate actual Cd-contaminated soil has economic feasibility. These findings provide evidence on behalf of the employment of nanoscale zero-valent iron nanocomposites for heavy metal remediation in contaminated soils.