The Remediation Strategy and Mechanism of Combined Passivation and Foliar Inhibition for Safe Rice Production in Red Paddy Soil Contaminated With Heavy Metals

To solve the problem of Cd in rice over the food-safe standard, the present eld study was carried out to explore the combined passivators (a mixture of quicklime (Q), polyacrylamide (A), sepiolite (S)) and Si-/Se- containing foliar inhibitors (Si or Se) at low (1) and high (2) application rates were added into the red paddy soil. After harvest the rice, the soil samples were collected to examine the soil properties, bacterial community, and the availability of heavy metals (Cd, Pb, and Cu) in soil. The rice samples were obtained to investigate the accumulation of heavy metals by rice. All of the treatments increased the soil's pH, but the treatments excluding QSe2 decreased the total P and soil organic matter (SOM), which was favourable for the immobilisation of heavy metals in red paddy soil by decreasing 14.29%-42.86% of available Cd, 10.18%-63.17% of available Pb, and 6.95%-36.81% of available Cu. With the increasing application rates, QAS signicantly decreased the heavy metals available because of the enhanced immobilisation, while QSi and QSe signicantly increased the heavy metals available because of the inhibited plant uptake. After remediation, QA1, QSi2, and QSe2 most effectively decreased the uptake Cd by rice in the present red paddy soil to solve the problem of Cd exceeding the threshold value according to the National Food Safety Standard of China (GB2762-2017). Additionally, the treatments, with the exception of Q1, QA1, QSi1, and QSi2, did not dramatically change the community structure of bacteria at the genus level in soil. Considering the safety and stability of soil, QSe2 was the primary recommendation for remediating Cd-contaminated red paddy soil. for

Heavy metals pollute about 2.35 × 10 12 m 2 of the world's cultivated land (Bermudez et al., 2012). According to China's rst national soil pollution survey, 19.4% of farmland sites exceeded the threshold of heavy metals in soil (MLRPRC 2014). The accumulation of heavy metals (i.e., cadmium (Cd), lead (Pb), and cuprum (Cu)) in rice due to soil pollution has become a topic of great public concern (Zhao et al. 2020; Hussain et al. 2020). This is especially true for red paddy soil as the strong acidity led to the high activity of heavy metals and easy uptake by rice ). Therefore, the safe utilisation of red soil contaminated by heavy metal is facing a great challenge (Chen et  Therefore, the aims of this study were to: (1) explore the effects of combined passivators and foliar Si-or Se-containing foliar inhibitors on the availability and uptake of heavy metals by rice in red paddy soil, (2) elucidate the effects of soil remediation practice on physicochemical properties and community structure of bacteria, and (3) identify the practical strategy for safe utilisation of red paddy soil contaminated with heavy metals. The results of this study will helpful provide insight into the immobilisation mechanism of heavy metals in red paddy soil by combined passivators and foliar Si-or Se-containing foliar inhibitors, which will be bene cial for the safe utilisation of such soil.
Jiangsu AgraForUm soil remediation Co., Ltd provided the foliar Si-or Se-inhibitors.
The soil used in screening components of combined passivators is the same as the soil in the eld experiment. The surface soil was collected from the heavy metal-contaminated site in Bishan Village, Taihe County, as described in Section 2.3.1. The tested soil was acid red soil. Basic physical and chemical properties: total phosphorus (TP): 462.67 mg·kg −1 , total nitrogen (TN): 2.47 g·kg −1 , soil organic matter (SOM): 24.37 g·kg −1 , cation exchange capacity (CEC): 10.40 cmol·kg −1 , and pH 4.58. The concentrations of total Cd, Pb, and Cu in soil were 0.28 mg·kg −1 , 36.33 mg·kg −1 , and 29.92 mg·kg −1 , respectively. The rocks, gravel, and branches were removed from soils. The air-dried soils were grounded to pass through a 2 mm aperture sieve. Soil pH, SOM, CEC, TN, and TP were analysed according to Lu (2000).

Laboratory Screening Experiment
The laboratory screening experiment was conducted to investigate the passivators' effect for immobilisation of heavy metal in red soil (Table S1). The bioavailability of Cd, Pb, and Cu in soils was determined in soil treated with different kinds of passivators (Fig. S1). Brie y, each passivator was applied to 100 g of soil at the ratio of 5% in a 200 mL round box and then well mixed. The mixture in each round box was ooded with distilled water. The weighing method was adopted to maintain the amount of distilled water in each round box. The samples without passivator were set as the control. After 7 d, all the samples were dried in a drying oven at 60°C. The dried soils were passed through a sieve (2 mm). There were three parallel samples for each treatment.

Field Experiment
This study started in July 2020 in red paddy soil in Bishan Village, Taihe County, Ji 'an City, Jiangxi Province, China (114°57′-115°20′E, 26°27′-26°59′N). In May 2020, the grid method was used for controlling the distribution of points and layout of surface soil sampling points: 0.67 ha of demonstration farmland was divided into grids of 25 m × 25 m. A mixed surface soil sample was taken from each grid with a depth of 0-0.2 m, and there were 11 surface soil samples in total. The concentration of Cd in the soil exceeded the standard in China (Soil environmental quality. Rick control standard for soil contamination of agricultural land. GB 15618-2018). Moreover, heavy metals were signi cantly enriched in the cultivated rice. The concentration of Cd in rice exceeded the threshold value according to the National Food Safety Standard of China (GB2762-2017) (data not provided).

Sampling And Determination
Five surface soils (0-20 cm) were taken from each grid and mixed into one sample. The three replicated soil samples were collected for each grid. All of the collected soil samples were air-dried at ambient temperature. The rocks, gravel, and branches were removed from the soils. The dried soil was sieved through a 2 mm nylon mesh for further analysis. At the ripening stage, the rice of each plot was collected with three replicates at the location corresponding to the soil sample. All rice samples were ground for further analysis.
To determine the total content of heavy metals in soil and rice, the samples were digested with HF-HNO 3 -

Data analysis
Statistical analyses were conducted using SPSS 23 software. The statistical difference between the results from different plots were analysed at the signi cance level of p < 0.05 using the one-way ANOVA and Duncan's at the 5% probability level analysis. Polychoric correlation coe cient analysis was used to compare the average values. Statistical signi cance was determined at 95% con dence interval with the signi cance level at p < 0.05. The "out-of-bag" method was used for Random Forest variable importance scoring. We used R (version 3.6.1) to determine the Random Forest importance ranking, which provided feedback for indicator (%IncMSE). The larger the calculated value, the greater its importance.

The bioconcentration factors (BCF), as de ned in the following equation (Zhang et al. 2017):
where is the concentration of heavy metal in rice grains, and is the concentration of total heavy metal in soil.

Effects of combined passivators and foliar inhibitors on soil properties
The application of combined passivators and foliar inhibitors into the red paddy soil generally signi cantly changed the soil's chemical properties ( Table 1). All of the combined passivators and foliar inhibitors increased the soil pH as compared with the control (p < 0.05 for most of the treatments). The signi cantly negative correlation between TP and soil pH (p < 0.05, Fig. 1) was most likely due to the increase in use e ciency of phosphorus by rice with increasing soil pH (Muhammad et al. 2020; Zhao et al. 2020). A signi cantly positive correlation was observed between the concentrations of SOM and TN in paddy soil (p < 0.001, Fig. 1). The consistent variation trend of TN and SOM was most likely due to the fact that TN in soil was largely related to the balance of accumulation and decomposition of SOM (Groffman et al. 2001).
Effects of combined passivators and foliar inhibitors on the availability of heavy metals in red paddy soil The combined passivators and foliar inhibitors decreased the available Cd, Pb, and Cu by 14.29%-42.86% (except for QSi2), 10.18%-63.17%, and 6.95%-36.81% (except for QSi2) as compared with the control, respectively (p < 0.05, with the exception of very few cases; Table 2). Others have also reported the decrease in the availability of heavy metals in soils with the incorporation of passivators (He et al. 2021).
For a given treatment, the trend of the concentration of available heavy metals was Pb > Cu > Cd, which was most likely related to their total contents in soil. From the perspective of application rates of combined passivators, the increase of Q1, QA1, and QS1 to Q2, QA2, and QS2 did not signi cantly change the available heavy metals in soil (p > 0.05, with the exception of Cd in Q1 and Q2; Table 2 and  Table S2). Therefore, when combined with the quicklime, polyacrylamide, and sepiolite (QAS), the signi cant decrease of available heavy metals in soil with the increase of QAS1 to QAS2 (p < 0.05; Table 2 and Table S2) was most likely due to the interaction of polyacrylamide and sepiolite as they enhanced the immobilisation of heavy metals. Unlike the changes in available heavy metals in QAStreated soils, the available heavy metals in the soils treated with QSi2 and QSe2 were signi cantly higher than those in the soils treated with QSi1 and QSe1 (p < 0.05; Table 2 and Table S2 The correlations between the available heavy metals in red paddy soil and soil properties are presented in Fig. 1. The soils treated with the combination of quicklime and foliar inhibitor were excluded in this correlation analysis due to its different mechanism on heavy metals. As seen in Fig.1, there were signi cant negative correlations between available heavy metals and soil pH (p < 0.05 for Cd and Pb, and These results indicate that the change in available heavy metals in soil was not only affected by soil pH, but also by other soil properties such as SOM and TP (Song et al. 2017).
The results from the rank of factors on available heavy metals in soil (Fig. 2) indicated that the dominant factors for available heavy metals were different depending on the types of the heavy metals. The rst two dominant factors regulating the available Cd, Pb, and Cu were TN and TP, CEC and TP, and TP and SOM. Evidently, TP was one of the key factors on the three available heavy metals studied.
Effects of combined passivators and foliar inhibitors on the uptake of heavy metals by rice For heavy metals-contaminated farmland soil, although the concentrations of total and available heavy metals in red paddy soil should not be ignored, the more critically important issue was to prevent and control rice's uptake of heavy metals in order to ensure food safety. From Table 2, the effect of the given combined passivators and foliar inhibitors on the uptake of heavy metals by rice was related to the types of heavy metals. Speci cally, the combined passivators and foliar inhibitors signi cantly decreased rice's uptake of Cd by 22%-93% as compared with the control (p < 0.05 excluding QS2), but generally had little in uence on rice's uptake of Pb (p > 0.05 excluding QA1 and QSe2). The variation of Cu in rice depended on the types and application rates of combined passivators and foliar inhibitors. For example, QA1, QSi1, QSi2, and QSe2 signi cantly decreased the uptake of Cu in rice (p < 0.05), while others had little in uence on (p > 0.05) or signi cantly increased (p < 0.05) the uptake of Cu by rice.
The signi cantly positive correlation between available Cd in red paddy soil treated with combined passivators (excluding foliar Si-or Se-inhibitors) and the uptake of Cd by rice (p < 0.01) indicated that the decrease of available Cd in red paddy soil studied was one of the effective approaches to control the  (Table S2 and Table S4). Similar uctuations were also observed after applying chicken or swine manure into soil (Wan et al. 2020).
With the increasing application dosage of QSi1 and QSe1 to QSi2 and QSe2, the uptake of heavy metals signi cantly decreased (p < 0.05 excluding Pb and Cu in soil treated with QSi). This observation was most likely due to the fact that Si and Se effectively inhibited the uptake of heavy metals because the increasing dosage of quicklime signi cantly increased their uptake (p < 0.05 with the exception of Pb; Table 2 The results from the rank of factors on uptake of heavy metals by rice (Fig. 2) indicates that the amount of available Cd in soil primary controlled its uptake. This nding further con rmed that the decrease of available Cd in red paddy soil was one of the effective approaches to control the uptake of Cd, as mentioned in Section 3.3. The main factors affecting the uptake of Pb and Cu were pH and CEC, respectively. The result indicates that soil chemical properties in the acidic red paddy soil studied played a more important role in affecting the uptake of Pb and Cu by rice as compared with the concentration of available heavy metals assessed by chemical extraction.
The variation trend of the BCF of heavy metals was similar to that of heavy metal uptake (Table S3 and Table 2). The BCF of Cd in the control was as high as 87.39%, indicating that the rice under natural conditions in this study showed a superb uptake of Cd. The combined passivators and foliar inhibitors signi cantly decreased the BCF of Cd (p < 0.05 excluding QS2), indicating that combined passivators and foliar inhibitors effectively inhibited the transport of Cd from soil to rice. However, the combined passivators and foliar inhibitors had little in uence on the BCF of Pb (p > 0.05 excluding QA1 and QSe2), indicating that the Pb translocation rate from soil to rice has not changed. This nding was different from a previous study that reported that the combination application of passivators and foliar inhibitors could suppress the formation of Fe/Mn plaques on rice roots and uptake by rice (Li et al. 2020b). The BCF of Cu increased signi cantly after applying Q2, QS, QAS, and QSe1 (p < 0.05), indicating the increased Cu translocation rate from soil to rice. Therefore, the combined passivators and foliar inhibitors should be used carefully for the safe utilisation of Cu-contaminated acidic red soil.

Microbial community response to the addition of combined passivators and foliar inhibitors into red paddy soil
The combined passivators and foliar inhibitors had little in uence on the richness (ACE and Chao index) and diversity (Shannon and Simpson index) of bacteria in red paddy soil studied (p > 0.05, Fig. S2). However, combined passivators and foliar inhibitors had in uence on the bacterial community at the phylum and genus level. The most abundant phylum of microbes found were Proteobacteria (15%-23%) and Chloro exi (14%-25%), followed by Acidobacteria (11%-17%) and Actinobacteria (8%-12%) (Fig. 3). This result is in agreement with Wang et al.  (Fig. 4). The clustering results from different treatments, indicating that the genera of bacteria in the CK were similar to those in the soils treated with most of the treatments, with the exception of Q1, QA1, QSi1, and QSi2, indicating that most of the treatments did not dramatically change the genus composition of bacteria in the red paddy soil studied. The high similarity of genera of bacteria in soil treated with Q1, QA1, QSi1, and QSi2, which was evidently separated with the other treatment groups. This result indicated that bacteria at the genus level were more sensitive to the quicklime, polyacrylamide, and foliar Si-inhibitors  The results in this study showed that applying combined passivators and foliar inhibitors (except for QS2) into the red paddy soil studied could reduce the concentrations of Cd in rice to permitted food-safe levels (< 0.2 mg kg −1 ), according to National Food Safety Standard of China (GB2762-2017) (MLRPRC 2014). Additionally, the application of combined passivators and foliar inhibitors did not lead to the other heavy metals being present in excess of standards. Among the combined passivators and foliar inhibitors, QA1, QSi2, and QSe2 showed the best effect for decrease in the uptake of Cd by rice in the red paddy soil studied. After taking the safety and stability of soil structure and function into consideration (Table 1 and Fig. 3), QSe2 (1875 kg·ha -1 Quicklime + 250 mL·mu -1 foliar selenium inhibitors) was the primary recommendation for remediating the heavy metals-contaminated red paddy soil, particularly for remediation of Cd-contaminated red paddy soil.
This study provided an e cient, inexpensive, and easy-to-operate strategy for remediating heavy metalscontaminated red soil in a paddy eld. However, different from traditional knowledge, it found that foliar inhibitors may also affect soil's physical-chemical properties and soil bacterial community. Some combined passivators decreased the concentrations of available Cu in soil, but increased Cu uptake in rice. In addition, the change of bacterial community at the genus levels may bene t the stabilisation of heavy metals in red paddy soil. In the future, it is necessary to determine the pathways or mechanisms by which foliar inhibitors affect soil chemistry and microbial structure. Future monitoring is necessary to explore the long-term effects of combined passivators on the safe utilisation of heavy metalscontaminated red paddy soil. Additionally, further studies are needed to investigate the change mechanism of soil physical and chemical properties and to elucidate the interaction mechanism between heavy metals and microorganisms.

Conclusion
Most of the combined passivators and foliar inhibitors increased soil pH, and generally decreased TN, TP, and SOM in paddy soil. Except for QSi2 (combined quicklime and foliar Si-inhibitors), the treatments decreased the soil Cd, Pb, and Cu available. Soil pH, TP, TN, and SOM played an important role in the immobilisation of heavy metals in red paddy soil. The treatments signi cantly decreased rice's uptake of Cd, which was mainly due to the decrease of available Cd in red paddy soil. Moreover, Si and Se effectively inhibited the uptake of heavy metals. Bacteria at the genus level were relatively sensitive to the quicklime, polyacrylamide, and foliar Si-inhibitors. The shift of bacterial community structure indirectly affected the soil available heavy metals. Except QS2, the concentrations of heavy metals in rice in other treatments were lower than the standards (GB2762-2017 National Food Safety Standard of China). Considering the safety and stability of soil, QSe2 (combined quicklime and foliar Se-inhibitors) was the primary recommendation for remediating the heavy metals-contaminated red paddy soil. Data availability The datasets used or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations
Ethical approval This study was approved by the ethics committee of Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, China.
Consent to participate All the listed authors agree.
Consent to publish Not applicable.
Con ict of interest We declare that we do not have any commercial or associative interest that represents a con ict of interest in connection with the work submitted.   CuRT stand for the total heavy metals in the rice. SeRT stand for the Se in the rice. * means p < 0.05, ** means p < 0.01, *** means p < 0.001.)  Heat map of the relative abundances of the predominant genera. PbRT and CuRT stand for the total heavy metals in the rice. SeRT stand for the Se in the rice. * means p < 0.05, ** means p < 0.01, *** means p < 0.001.)

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.