Introduction

During the past few decades, soil contamination by heavy metals has become a serious issue in agriculture, mainly due to anthropogenic activities, such as mining, smelting operations, sewage irrigation, and fertilizer application (Rai et al. 2019; Raj and Maiti 2020). Regarding heavy metals, particularly cadmium (Cd) and arsenic (As, a metalloid but usually recognized as a heavy metal) are extremely hazardous pollutants in agricultural soils, giving rise to suppression of crop growth, excessive Cd or As in cereals, and threats to human health through food consumption (Rai et al. 2019; Raj and Maiti 2020). Rice (Oryza sativa L.) is one of the foremost cereal crops over the globe, but has a high ability to uptake Cd and As from soils and thus has become a major dietary source of Cd and As (Irshad et al. 2020). Geochemical behaviors of Cd and As vary greatly under paddy field condition, generating a major risk for rice cultivation (Irshad et al. 2020). To ensure safe cereal production, the remediation of paddy soils contaminated by Cd and As is imminently needed. One widely employed technology is in situ chemical stabilization of heavy metals by adding soil amendments (Kumpiene et al. 2019).

Steel slags, consisting of main constituents of silicon (Si), calcium (Ca), phosphorus (P), iron (Fe), and manganese (Mn), are low-cost, abundant, and alkaline byproducts from steel mills (Navarro et al. 2010; Guo et al. 2018). In some countries, steel slag has been maturely used as silicate fertilizer, phosphate fertilizer, and soil modification in agriculture (Guo et al. 2018). Steel slag application improves soil quality (León-Romero et al. 2018), increases soil nutrient availability and rice yield (Wang et al. 2015a; He et al. 2017; Wang et al. 2018a), and reduces methane emission in paddy fields (Wang et al. 2015a; Wang et al. 2018b). Further, steel slag amendment can immobilize heavy metals (e.g., Cd and lead) in soil and decrease their concentrations in crops (Kim et al. 2012; Ning et al. 2016; He et al. 2016; Hu et al. 2019), but the amendment may have positive or negative impacts on As uptake by plants (Nejad et al. 2017; León-Romero et al. 2018). Nevertheless, information is lacking regarding steel slag for the remediation of co-contaminated paddy soils with Cd and As, and the possible mechanisms behind steel slag effects on the accumulation of Cd and As in crops are poorly understood.

Heavy metals uptake by plants depends mainly on the solubility and availability of these metals in soil (Zeng et al. 2011). The heavy metals measured in soil pore-water denote the most soluble fractions in soil for plant uptake (Beesley et al. 2010; Moreno-Jiménez et al. 2011; Concas et al. 2015) and are good indicators to assess the remediation efficiency of soil amendments (Zheng et al. 2012; Beesley et al. 2014). Schemes of soil amendments may influence the solubility of Cd and As in soil by the modification of chemical properties altering soil pH and binding forms of metals in soil solution (Beesley et al. 2014). Solubility of heavy metals is also affected by the application of beneficial nutrients such as Si and P, owing to their impacts on adsorption of metals onto soil and metal bioavailability for crop uptake (Sarwar et al. 2010). It is possible that alkaline steel slag containing Si- and P-rich compounds can change the aforementioned soil factors, therefore affecting the accumulation of Cd and As in rice. In addition, iron plaque is generally formed on rice root surfaces and can modify the uptake and accumulation of metal(loid)s in rice plants (Liu et al. 2004; Wang et al. 2013; Cheng et al. 2014). The formation of iron plaque and its adsorption capacity on metals are affected by soil conditions and soil amendments (Zheng et al. 2012). As reported, the biochar amendments increased iron plaque formation on the surfaces of rice roots and sequestrated more Cd and As in iron plaque (Irshad et al. 2020), whereas the Cd concentrations in iron plaque decreased (Zheng et al. 2015). It is hypothesized that supplementing paddy soils with steel slag may affect iron plaque formation and thereby the uptake of Cd and As by rice; however, this potential effects are unclear.

Based on the above observations and hypothesis, this work aimed to study the effects of steel slag on the solubility of Cd and As in soil and their accumulation and translocation within rice plants grown in a historically co-contaminated paddy field with Cd and As, and to explore the possible mechanisms involved. Paddy field (rhizobag) experiments were carried out to explore the effects by determining soil pH; soluble Cd, As, Si, and P in soil pore-water; concentrations of Cd, As, and Fe in iron plaque; concentrations of Cd and As in rice tissues (roots, straw, and brown rice); and translocation of Cd and As from roots to aerial parts. Rice biomass and root antioxidant enzyme activities were also measured.

Materials and methods

Experimental location, steel slag, and rice variety

The experimental site was established at a paddy field in Shangba village (24°27′N; 113°48′E) close to Dabaoshan mine, Guangdong Province, southern China. The area has a subtropical humid climate with an average annual rainfall of 1762 mm and an average annual temperature of 20.3 °C. Dabaoshan mine is a cluster of opencast multi-metal mines initiating operation in the 1960s and is still in operation. The studied paddy field has been chronically contaminated due to the acid mine drainage from Dabaoshan mine. The total concentrations of Cd and As in the paddy soil (Table 1) were, respectively, 8.2-fold and 2.6-fold above the Soil Environmental Quality – Risk Control Standards for Soil Contamination of Agricultural Land (Cd 0.3 mg kg−1 and As 30 mg kg−1, in GB 15618–2018 of China). Thus, the soil markedly exceeded the national risk control levels for Cd and As in soil and was considered relatively highly contaminated. In this study, steel slag powders through a 0.15-mm sieve were obtained from Shaoguan Steel Group Company of Guangdong Province (He et al. 2016). The slag was strongly alkaline and has very low concentrations of Cd and As (Table 1), and could be considered safe for use in agricultural soils referring to the maximum allowable content of Cd and As in the Control Standards of Pollutants in Fly Ash for Agricultural Use (Cd 5 mg kg−1 and As 75 mg kg−1, in GB 8173–87 of China). Seeds of rice cultivar (Oryza sativa cv. Tianyou 122) were from the Rice Research Institute of Guangdong Academy of Agricultural Sciences and were used in field experiments.

Table 1 Properties of the paddy soil (mean ± SE, n = 5) and steel slag (mean ± SE, n = 4)
Full size table

Paddy field (rhizobag) experiments

A total of 12 experimental plots were set up with each plot having an area of 6 m2 (2 m × 3 m), 30-cm ridges around each plot, and 20-cm spacing between adjacent plots. The plots were randomly divided into three treatments, each with four replicates. The treatments included control (without steel slag) and steel slag amendments with the addition (low and high dosages) of 2.0 t slag ha−1 (SS1) and 4.0 t slag ha−1 (SS2), according to our previous study (He et al. 2016). Steel slag was added into each plot and mixed thoroughly with the soils of a 15-cm tillage layer. Subsequently, about 1.0 kg of the mixed soils in each plot were added to a cylindrical rhizobag (made of 30-μm nylon mesh, 12-cm diam, 15-cm height) that was designed to differentiate rhizosphere from non-rhizosphere zones (Cheng et al. 2014; Wang et al. 2015b). The interface between soils and plant roots is the rhizosphere that is the most active site for direct material exchange in soil-plant systems (Wang et al. 2015b). Thus, the effects of steel slag on paddy soils, rice plants, and their possible relationships may be availably explored using rhizosphere bag. Four rhizobags were placed into the central zone of each plot, prior to transplanting 20-day-old seedlings which were first cultivated in a seedbed at an unpolluted field. Two rice seedlings were transplanted to a rhizobag that was gently taken from the field at harvested stages, and the transplantation of the rest rice seedlings followed the conventional way in each plot. Rice plants were cultivated for 90 days and were harvested at heading and mature stages. All experimented plots followed the same agricultural managements, such as irrigation with clean water, fertilization, and weeding (He et al. 2016).

Sampling and analytical methods

Samples of rice plants harvested were separated into roots and straw at the heading stage, and roots, straw, and grains at the mature stage. Straw and grains were oven-dried at 50 °C until a constant weight for biomass. Roots were taken from rhizobag soils by gentle sieving for further measurements. The activities of antioxidant enzymes (catalase (CAT), superoxide dismutase (SOD), and ascorbate peroxidase (APX)) of fresh roots gathered at the heading stage were determined as reported by Liu et al. (2013). Iron plaque on root surfaces was extracted using dithionite-citrate-bicarbonate (DCB) solution (Otte et al. 1989; Liu et al. 2004). After washing with deionized water, rice roots were incubated in 40 ml of DCB solution for 60 min, and then were rinsed three times with deionized water. Rinsed water was transferred to the DCB extracts and the final solution was made up to 50 ml for analysis of Cd, As, and Fe in iron plaque. Subsequently, the dried samples of roots, straw, and grains were pretreated and analyzed for Cd and As according to the methods of Wang et al. (2013).

Soil samples were collected from the rhizobags and were air-dried. Soil pH was measured in a slurry with a water–solid ratio of 2.5:1 using a pH meter (pH 510, Eutech Instruments, Singapore). Soil pore-water was collected by in situ pore-water samplers (Rhizosphere Research Products, the Netherlands) inserted into the base of the rhizobags, referring to the method described by Kidd et al. (2007) and Zheng et al. (2012). The pore-water was sampled at both heading and mature stages and concentrations of Cd, As, Si, and P in pore-water were determined. Elemental concentrations in samples were measured using graphite furnace atomic absorption spectrophotometer (GFAAS, Hitachi Z-2000, Japan) for Cd, atomic fluorescence spectrometry (AFS, Beijing, Jitian Instrument Co., Ltd.) for As, and inductively coupled plasma optical emission spectrometry (ICP-OES, Optima 2000 DV, Perkin Elmer, USA) for Fe, Si, and P. Blanks, soil standard material (GBW-07435), and rice standard material (GBW-10045) (China Standard Materials Research Center, Beijing, China) were performed for quality control.

Statistical analysis

Translocation factors for Cd or As from roots to aerial parts were calculated as follows: TF = Cd or As concentrations in aerial parts/roots. Statistical analyses were performed using SPSS (version 18.0) software for Windows. Results were presented as arithmetic means with standard error attached, and means were compared using one-way variance (ANOVA) followed by the least significant difference (LSD) test at 5% probability level.

Results

Steel slag effects on soil pH and pore-water Cd, As, Si, and P

Steel slag amendments of 2.0 t ha−1 (SS1) and 4.0 t ha−1 (SS2) showed beneficial effects on pH and soluble Cd (but not As), Si, and P in rhizosphere soils at both the heading and mature stages of rice, and the effects were greater in SS2 treatment than those in SS1 treatment (Fig. 1). Compared with the unamended controls, both treatments decreased soluble Cd concentrations in pore-water, and especially SS2 treatment resulted in significant (P < 0.05) reductions by 45% and 34% at the heading and mature stages, respectively (Fig. 1). In contrast to the trend for Cd, both treatments enhanced As concentrations in pore-water, with SS2 affecting significant (P < 0.05) increases by up to 2.5-fold at both stages (Fig. 1).

Fig. 1
figure 1

Rhizosphere soil pH and pore-water concentrations of Cd, As, Si, and P in paddy field experiments with and without steel slag (mean ± SE, n = 4). SS1: steel slag (2.0 t ha−1), SS2: steel slag (4.0 t ha−1). Different letters within the same parameter and the same stage indicate significant difference (P < 0.05) between the treatments

Full size image

In addition, both SS1 and SS2 treatments promoted significant increases in soil pH in excess of 1.1–2.5 units (P < 0.05), and also improved soluble concentrations of Si and P in pore-water at both stages, particularly for SS2 causing significant (P < 0.05) increases in pore-water Si and P by 1.9–2.7-fold (Fig. 1).

Steel slag effects on Cd, As, and Fe in iron plaque

Compared with the controls, the formation of iron plaque (DCB-Fe) on rice root surfaces was not significantly increased by steel slag amendments at both the heading and mature stages (Fig. 2). The influence of steel slag varied on concentrations of Cd and As in iron plaque. Both treatments significantly (P < 0.05) decreased Cd concentrations in iron plaque (DCB-Cd) at both stages, while the concentrations of As in iron plaque (DCB-As) increased markedly (P < 0.05) at the heading stage (Fig. 2).

Fig. 2
figure 2

Concentrations of Cd, As, and Fe in iron plaque on rice root surfaces in paddy field experiments with and without steel slag (mean ± SE, n = 4). SS1: steel slag (2.0 t ha−1), SS2: steel slag (4.0 t ha−1). Different letters within the same parameter and the same stage indicate significant difference (P < 0.05) between the treatments

Full size image

Steel slag effects on Cd and As accumulation and translocation in rice

As expected, the Cd concentrations in rice tissues (roots, straw, and brown rice) decreased after steel slag amendments at the heading and mature stages, and the reductions were more in SS2 treatment than those in SS1 treatment (Table 2). Compared with the controls, SS2 treatment significantly (P < 0.05) reduced the concentrations of Cd in rice tissues by 48–78% at both stages, and the concentrations of Cd in brown rice were lower than 0.2 mg kg−1 (the limit of the Food Quality Standard GB2762–2017 of China). Furthermore, both treatments decreased the translocation factors of Cd from roots to aerial parts at both stages, and significant differences (P < 0.05) were observed at the heading stage (Table 2).

Table 2 Concentrations (conc., mg kg−1) of Cd and As in rice tissues (roots, straw and brown rice) and their translocation factors from roots to aerial parts in paddy field experiments with and without steel slag (mean ± SE, n = 4)
Full size table

In contrast to the Cd concentrations, steel slag amendments increased the concentrations of As in rice tissues (roots, straw, and brown rice) by 13–38% at the heading and mature stages, and notably, SS2 treatment resulted in significant (P < 0.05) increments (29–38%) in roots at both stages and in straw at the heading stage, compared with the controls (Table 2). No significant change in As translocation from roots to aerial parts existed after steel slag amendments (Table 2).

Steel slag effects on rice biomass and root antioxidant enzyme activity

Compared with the controls, steel slag amendments enhanced the dry biomass of rice tissues (roots, straw, and grains) at both the heading and mature stages, and significant (P < 0.05) increases in plant biomass (roots, straw, and grains) by SS2 treatment ranged from 24 to 40% (Table 3). Besides, steel slag amendments slightly increased the activities of the antioxidant enzyme SOD in rice roots, but the activities of CAT and APX increased significantly (P < 0.05) by SS2 treatment (Table 3).

Table 3 Biomass of rice tissues (roots, straw and grains) (kg plot−1 dry weight) and activities of antioxidant enzymes, catalase (CAT, u g−1 min), ascorbate peroxidase (APX, u g−1 min), and superoxide dismutase (SOD, u g−1) of fresh rice roots in paddy field experiments with and without steel slag (mean ± SE, n = 4)
Full size table

Correlations between the parameters

The linear regression analysis revealed that, after steel slag amendments, the concentration of Cd in pore-water was negatively correlated with soil pH (R = − 0.993; P < 0.01), pore-water Si concentrations (R = − 0.953; P < 0.01), and pore-water P concentrations (R = − 0.876; P < 0.05), but pore-water As was positively correlated with soil pH (R = 0.932; P < 0.01), pore-water Si (R = 0.983; P < 0.01), and pore-water P (R = 0.911; P < 0.05) (Table 4). Significantly positive correlations were also found between pore-water Cd and Cd concentrations in roots (R = 0.918; P < 0.01) and between pore-water As and As concentrations in roots (R = 0.950; P < 0.01) (Table 4).

Table 4 Correlation coefficients (R) between concentrations (μg L−1) of Cd or As in pore-water and soil pH, pore-water Si concentrations (mg L−1), pore-water P concentrations (μg L−1) and concentrations (mg kg−1) of Cd or As in rice roots, respectively (n = 6; *P < 0.05, **P < 0.01)
Full size table

Discussion

Excessive accumulation of Cd and As in crops has become a pressing concern. The present study demonstrated that steel slag amendment significantly increased rice production and decreased Cd concentrations in rice tissues and in iron plaque, though enhanced rice concentrations of As (Table 2), being a new suggestion that steel slag seemed impossible to simultaneously mitigate the accumulation of Cd and As in rice plants. To our knowledge, it is the first to investigate the effects of steel slag on simultaneous alteration of Cd and As in iron plaque and their accumulation in rice plants grown in a historically co-contaminated paddy field with Cd and As. Most previous studies in steel slag application have been limited to glasshouse or laboratory conditions, and the results may therefore not directly predict the field behavior. Thus, a discrimination of mechanisms by steel slag controlling the mobility of Cd and As from soils to rice plants enables the benefit of steel slag application under field environment.

The contrasting effects of steel slag on As and Cd in rice plants may be related to the amendment impact on soil geochemical properties. Steel slag amendment clearly decreased soluble Cd concentrations of pore-water in rhizosphere soils but increased that of As, compared with the controls (Fig. 1). The changes of Cd and As in pore-water are in consistence with the alkaline biochar study by Zheng et al. (2012) and Beesley et al. (2014). Possibilities below can be envisaged to explain the decrease of Cd solubility in soil by steel slag amendments. It is generally recognized that soil pH is a key factor regulating Cd concentrations in soil solution, and an elevated soil pH can reduce soluble Cd by increasing adsorption or precipitation (Wang et al. 2019). The strongly alkaline nature gives steel slag a high capacity to increase pH in acid environment (Navarro et al. 2010), likely leading to the precipitation of soil Cd as cadmium hydroxides. Besides the pH, the amount of high Si and P in steel slag enhanced the concentrations of Si and P in pore-water (Fig. 1). Wang et al. (2018a) also found that steel slag application increased the extractable Si and P in paddy soils. Improving Si and P nutrition could decrease the solubility and availability of Cd in soil by inducing the formation of Cd–Si co-precipitation and insoluble Cd phosphate (Sarwar et al. 2010). Lu et al. (2014) reported that Si additions restrained Cd uptake by crops, largely due to the alteration in soil chemistry, e.g., increasing soil pH and Cd adsorption, and reducing Cd competitiveness for crop uptake. Similarly, previous researches revealed that steel slag amendment decreased the extractable soil Cd related to the improvement of pH and extractable Si in pot experiments (Ning et al. 2016; He et al. 2017). Nevertheless, it was observed that the soluble P raised by steel slag might be another factor stimulating soil Cd immobilization in the present experiment (Fig. 1 and Table 4). Altogether, the mechanisms underlying the alteration of soluble soil Cd caused by steel slag are likely the increases of soil pH and soluble Si and P in pore-water (Fig. 1), as confirmed by the regression analysis in this study (Table 4).

Regarding As, unlike Cd, the present regression analysis confirmed the positive relationships of pore-water As concentrations with soil pH and pore-water Si and P concentrations, after steel slag amendments (Table 4). Analogously, the biochar and compost amendments promoted the solubilization of As to pore-water associated with increased pH and soluble P in soil (Beesley et al. 2014). It was explained that the pH increases by soil amendments decreased the As adsorption on soil minerals, and phosphate competed with As for binding sites in soil (Fleming et al. 2013; Beesley et al. 2014). Further, Si additions increased the As concentrations in soil solution due to the Si competition for As adsorption sites of soil solids (Lee et al. 2014). In general, alkaline materials, such as lime, fly ash, and biochar, may be undesirable for remediating As-contaminated soils as they cause higher pH with higher leaching and mobility of As (Kumpiene et al. 2008, 2019). Consequently, steel slag enhanced the solubility of soil As possibly by a combined effects of increased soil pH and pore-water Si and P (Fig. 1).

Generally, the solubility and availability of heavy metals in soil are the primary factors affecting the uptake of these metals by plant roots (Zeng et al. 2011). In this study, steel slag amendments decreased Cd concentrations and increased As concentrations in rice tissues (roots, straw, and brown rice) (Table 2), and the regression analysis revealed strongly positive correlations of Cd and As in roots with their concentrations in pore-water, respectively (Table 4). In agreement with our results, the declined concentrations of Cd in rice plants upon the addition of steel slag were observed, causing the decrease in extractable soil Cd (Ning et al. 2016; He et al. 2017). León-Romero et al. (2018) found that steel slag application promoted As uptake by roots and stems of Arabidopsis thaliana in pot experiments, regarding the possible reasons that were the alterations in As absorption and phosphate in soil. However, besides the pore-water P, the present results implied that soil pH and pore-water Si might generate evident impacts on the increases in the solubility and accumulation of As induced by steel slag amendments (Table 4). In addition, steel slag amendments decreased Cd but enhanced As in iron plaque, though the formation of iron plaque was not markedly increased by the amendments (Fig. 2). It is noted that iron plaque uptake of metals depends much on the bioavailability of metals in plant rhizosphere and the amounts of iron plaque, despite the wide recognition that iron plaque has a capacity to retain metals (Cheng et al. 2014). Similar to our results, Zheng et al. (2012, 2015) reported that the biochar amendments induced less Cd in iron plaque compared with the controls, owing to the decreased Cd in pore-water, thereby generating less transfer to the root surfaces from soil solutions. In contrast, the elevated pore-water As by steel slag amendments may have affected As onto the iron plaque and its concentrations in the plaque (Figs. 1 and 2). However, Yu et al. (2017) found that iron-based amendments reduced As dissolution in rhizosphere soils and increased iron plaques, contributing to decrease As concentrations in rice plants in pot experiments. Rice As accumulation varied among soil amendment systems, likely owing to the differences of the amendments used and their effects on As bioavailability in soil, iron plaque formation, and As mobility within rice plants. Thus, the formation of iron plaque and the plaque effects on rice As accumulation varied possibly. Furthermore, Cd translocation from roots to aerial parts was restrained by steel slag additions (Table 2). One possible explanation is the increases of Si concentrations and its co-precipitation with Cd in rice tissues (Ning et al. 2016). However, no significant alteration of As translocation by steel slag amendments may be attributed to the complicated pathways of As transport within rice plants, involving mechanisms of arsenate reduction, arsenite transport, As complexation and sequestration, As methylation, and As alteration in xylem and phloem (Zhao et al. 2010). More detailed investigations are required to understand the role of steel slag for the transfer process of As in crop plants.

Crop growth is also an important factor for evaluating the efficiency of soil amendment in agriculture lands (Kumpiene et al. 2019). Here steel slag amendments clearly improved the biomass of rice plants (roots, straw, and grains) and the activities of root antioxidant enzymes (Table 3). As reported, steel slag additions raised soil pH and resulted in a more neutral soil environment form the acidic one, benefiting rice growth (He et al. 2016). Besides, the increased solubility of nutrients like Si and P by steel slag additions contributed to rice biomass (Wang et al. 2015a, 2018a) and root antioxidant enzyme activities (Sarwar et al. 2010). Overall, the application of steel slag can be beneficial for grain production and Cd (not As) reduction in rice under field condition.

Conclusions

This study demonstrated that steel slag amendment effectively increased rice production and decreased the concentrations of Cd in rice tissues (roots, straw, and brown rice) and in iron plaque, though enhanced those of As, in a historically co-contaminated paddy field with Cd and As. The possible mechanism behind the effects of steel slag on Cd and As in rice is likely the slag decreasing soluble Cd and increasing soluble As in pore-water, related to significantly increased soil pH and soluble Si and P in pore-water, and suppressing Cd translocation from roots to aerial parts. Besides, the increments in soil pH and soluble nutrients (Si and P) cloud conduce to the improvement of rice yield and root antioxidant enzymes. These results suggest that steel slag amendment represents a favorable potential for the remediation of Cd-contaminated paddy soils, but seems undesirable for Cd and As co-contamination. Further investigations of field trials in multi-element contaminated sites are needed to evaluate the long-term effectiveness of the amendment.