1 Introduction

Heavy metal-contaminated soil is a worldwide problem that urgently needs to be solved, especially in China where more than 10 % of farmlands have been contaminated by heavy metals and no fewer than 11 provinces have been affected, especially south of the Yangtze River (Tencent 2012). According to data from the Bulletin of Soil Pollution Situation Investigation of People’s Republic of China (MEPC 2014), agricultural soil has been polluted mainly due to anthropomorphic activities (such as industry, mining, and agriculture) and has resulted in soil levels above the standard for Cd, Pb, and Cu in 7.0, 1.5, and 2.1 % of the cases, respectively. Unlike organic contaminants, heavy metals in soil are non-biodegradable and, therefore, persist in soil for a long time, leading to long-term (chronic) effects on the soil environment (Bolan et al. 2003). Heavy metals in soil can threat people’s health either by accidental soil ingestion, by breathing contaminated soil dust particles or by the ingestion of polluted drinking water or farm products associated with contaminated soil. Therefore, it is necessary to implement appropriate soil remediation practices to reduce the bioavailability of metals in soils in order to protect human health.

Conventional soil-remediation technologies, such as excavation, transport, soil replacement, and landfilling of contaminated soil, are effective, but may be too expensive for the remediation of large areas of contaminated agriculture fields. The agricultural-based approach of in situ stabilization by applying soil amendments was recognized as a realistic choice in the case of vast contaminated farmlands (Mulligan et al. 2001). While in situ stabilization treatments do not reduce heavy metal concentrations in soil, they do enhance the soil mechanisms that controlling heavy metal mobility and bioavailability. It has been reported that the addition of various amendments (e.g., lime, hydroxyapatite, silicon materials) to soil decreased the mobility and bioavailability of heavy metals and reduced their toxic potential (Bolan et al. 2003; Vieira da Cunha and Araujo do Nascimento 2009; Padmavathiamma and Li 2010).

In recent years, many studies have focused on using industrial wastes and by-products as amendments in order to remediate the soil and, at the same time, putting these materials to good use. A number of materials, such as red mud, steel shot, furnace slag, sugarbeet lime, and flue gasses desulfurization (FGD) residue (Friesl et al. 2004; Garrido et al. 2005; Lee et al. 2011a) have been used, and the remediation effects have varied depending on the particular metals, soil properties, and experimental conditions. Usually, the industrial wastes applied as amendments were simply ground up or sieved, at best before use and lacked any specific processing designed to achieve better remediation results. These materials did not reduce heavy metal toxicity efficiently, and sometimes, large doses were needed to achieve the desired level of remediation (Stehouwer et al. 1995; Lee et al. 2011a; Ahmad et al. 2012), which often had negative influences on soil properties, such as compaction, overliming and basification. On the other hand, the amendments can enhance the amount of some mineral nutrients (K, Ca, Mg, Si, and/or P) that contribute to the restoration of contaminated farmland. Only a few studies have reported on the effects of the added mineral nutrients on the mitigation of toxic heavy metals and their effect on plant growth.

In this work, three modified industrial by-products (modified low-grade phosphate rock (MP), modified flue gasses desulfurization (MF), and a mixture of MF and phosphoric acid (MFP)) were applied to heavy metal-contaminated soil collected from a farmland near a copper and iron mine. The specific objectives were to (i) examine the efficiency of the three modified industrial by-products on heavy metals stabilization in polluted soil; (ii) estimate the effects of the amendments on the improvement of soil quality (acidity amelioration and increase in mineral nutrients) and plant growth; and (iii) analyze the correlations of soil and plant parameters in order to investigate the potential mechanism(s) of reducing heavy metals toxicity.

2 Materials and Methods

2.1 Sampling and Characterization

Top soil (<20 cm depth) was collected from a farmland adjacent to Dabaoshan copper and iron mine, Shaoguan City, Guangdong province, People’s Republic of China (24.47 N, 113.8 E). The soil sample was air-dried and sieved at <2 mm in preparation for determining its physiochemical properties. Soil organic matter (OM) was determined according to the Walkley–Black procedure (Nelson and Sommers 1982). Soil pH was measured using 1:2.5 (by weight) soil:H2O mixture. Available K, Ca, and Mg in the soil was extracted with 1 M ammonium acetate (Ok et al. 2007) and analyzed using atomic absorption spectrometry (Shimadzu, AA6300C, Japan). For available P, soil was extracted with 0.5 M NaHCO3, and P in the extracts was determined following an ascorbic acid blue color method (Watanabe and Olsen 1965). Soil available silicon was tested by the method of Acquaye and Tinsley (1965).

The digestion of soil samples was performed using the HNO3, HF, and HClO4 hot block digestion procedure, and heavy metals were measured by graphite furnace atomic absorption spectrometry (GFAAS) (Shimadzu, AA6300C, Japan). Some properties of the soil are presented in Table 1.

Table 1 Basic properties of the studied soil (mean values of three replicates)
Full size table

2.2 Soil Amendments

The amendments used in this experiment were three modified industrial by-products. The first amendment was a modified low-grade phosphate rock (MP), prepared by calcining a mixture of low-grade phosphate rock and soda dregs at an appropriate ratio. The low-grade phosphate rock was collected from a phosphate ore-dressing plant in Zhanjiang, People’s Republic of China, and the soda dregs, discharged from the ammonia-soda process, were derived from the South China Soda Ash Manufacturing Co., Ltd, Guangzhou, People’s Republic of China. The low-grade phosphate rock and soda dregs were finely ground and sieved before being mixed in an appropriate P2O5/CaSO4 molar ratio. Then, the mixture of the two industrial by-products was calcined for 1 h at 1000 °C to activate the inactive apatite and SiO2 in the low-grade phosphate rock. A nutrient amendment with available P, Si, Ca, and Mg was obtained with the above process.

The second amendment was a modified flue gasses desulfurization residue (MF) prepared by calcining a mixture of FGD residue and potassium feldspar. The preparation process has been reported in detail by Shi et al. (2011). The FGD residue and potassium feldspar were finely ground and mixed together based on the ratios in the following equation:

$$ 2{\mathrm{K}\mathrm{AlSi}}_3{\mathrm{O}}_8+{\mathrm{Ca}\mathrm{SO}}_3+14{\mathrm{Ca}\mathrm{CO}}_3={\mathrm{K}}_2{\mathrm{SO}}_4+{\mathrm{Ca}}_3{\mathrm{Al}}_2{\mathrm{O}}_6+6{\mathrm{Ca}}_2{\mathrm{SiO}}_4+14{\mathrm{CO}}_2 $$

The mixture of the two materials was calcined for 1 h at a temperature of 1000 °C in a rotary furnace, resulting in an amendment with available K, Ca, Si, and Mg. The third amendment (MFP) was prepared as a mixture of phosphoric acid and MF at 5 % (w/w). This material could provide P in the amelioration process which was absent in the MF treatment alone. Some properties of the amendments are given in Table 2.

Table 2 Characterizations of the amendments (mean values and standard deviations of three replicates)
Full size table

2.3 Experiment Setup

The amendments were applied to soils at doses of 0.5, 1, 2, and 4 ‰ by weight and thoroughly mixed with the soil to obtain homogeneity. Non-amended soils were used as a control group. Then, the soils were placed plastic pots (3 kg capacity, 19 cm diameter, and 17 cm height, three replicates per treatment). Twenty pakchoi seeds were sown in each pot, which were culled to five plants 2 weeks after germination. Soil moisture was maintained at 75 % field capacity by adding water to compensate for water loss. The experiment was conducted under controlled conditions (temperature 20–30 °C, relative humidity 60–70 %) with daily watering.

2.4 Plant Analysis

After 6 weeks, pakchoi was harvested with shoot (leaves) separated from root. The growth of pakchoi in each treatment was evaluated by measuring shoot height and, shoot and root mass. After being washed thoroughly with deionized water, the leaves and roots were dried at 70 °C for 48 h. The digestion of plant tissues was performed using the HNO3/H2O2 hot block digestion procedure (EPA 1986). Heavy metals (Cd, Pb, and Cu) in the digested solutions were determined using GFAAS (Shimadzu, AA6300C, Japan).

2.5 Five-Step Sequential Extraction

Changes of the fractions of heavy metals were determined using the procedure described by Tessier et al. (1979). Heavy metals were separated into five operationally defined fractions. First, a soil sample was extracted with 8 mL 1 M NaOAc at pH 8.2 for 1 h, producing F1, the exchangeable fraction. Second, the fraction bound to soil carbonates was extracted from the residue of first step with 1 M NaOAc at pH 5, producing F2, the carbonate-bound fraction. Third, the fraction bound to Fe and Mn oxides was extracted with 0.04 M hydroxylamine hydrochloride in 25 % (v/v) HOAc, producing F3, the Fe-Mn oxide fraction. Fourth, the fraction bound to organic matter was obtained after heating the soil residue from the previous step in 0.02 M HNO3 and 30 % H2O2 for 3 h, followed by extraction with 3.2 M NH4OAc for 30 min, producing F4, the fraction bound to organic matter. The fifth and final residual fraction was obtained by digesting the residual of soil samples with HF-HClO4, producing F5, the residual fraction. Heavy metals in the extraction solutions were determined with GFAAS (Shimadzu, AA6300C, Japan).

2.6 Statistical Analysis

All the data were statistically analyzed using one-way ANOVA at a significance level of P < 0.05 (marked by * in this paper) and P < 0.01(marked by ** in this paper) with SPSS 17.0 software@. The Duncan test was used to detect the significant differences between the means of the different treatments.

3 Results and Discussion

3.1 Effect of the Amendments on Soil pH

Figure 1 presents the change of soil pH after being amended with MP, MF, and MFP. Owing to the strong alkalinity of the amendments (Table 2), soil pH increased by 0.13–0.88 units with MP, 0.32–1.26 units with MF and 0.11–1.16 units with MFP when compared with the control. The increase of soil pH was consistent with the increasing rates of amendments applied.

Fig. 1
figure 1

Effect of the amendments on soil pH

Full size image

Unreacted CaO, MgO, and new substances of the amendments, such as MgSiO3, Ca2MgSi2O7, CaMgSiO4, and some other amorphous Ca and/or Mg compounds, may react with H+ ions in the soil solution (pH = 3.83) which increased the soil pH directly. In this study, pH showed significantly positive correlations with available Ca and Mg (r = 0.776** and 0.862**, respectively) which were similar to the findings of Ok, who found that increased available Ca led to the release of OH− ions in the soil solution (Ok et al. 2011). Also, increased availability of Ca2+ and Mg2+ in the amended soil (Table 3) reduced the exchangeable acidity (Noble and Hurney 2000), which improved soil resistance to acidification.

Table 3 Effect of amendments on mineral nutrients of soil
Full size table

3.2 Effect of the Amendments on Mineral Nutrients in Soil

As presented in Table 3, almost all of the three amendments resulted in improvement of available mineral nutrients; i.e., K, Ca, Mg, Si and/or P.

In the MP treatments, available K decreased significantly compared to the control (Table 3), possibly due to the better growth of pakchoi (Fig. 7) which assimilated much more K than the control but without extra K to replenish the adsorption loss (Table 2). In contrast, for the MF and MFP treatments, since the MF and MFP could provide adequate K that the pakchoi needed (Table 2), especially in the 2 and 4 ‰ mixtures, soil available K increased significantly (p < 0.05) compared to the control. As all the amendments are rich in available Ca, Mg and Si, especially MF and MFP, the addition of the amendments increased soil Ca, Mg, and Si content as a whole and the increase reached statistical significance (p < 0.05) in the 2 and 4 ‰ mixtures. The application of P-rich amendment (MP) enhanced soil available P significantly (p < 0.05), but there were no significant differences observed among MF treatments which could provide less available P (Table 2). Although the MFP treatment increased soil available P, the increases did not reach significant level. These findings indicate that soil nutrients status was improved in response to the application of the amendments.

3.3 Effect of the Amendments on Heavy Metal Distribution: Sequential Extraction

In this study, sequential extraction was used to investigate the distribution of metals among different soil pools. The relative distributions of Cd, Pb, and Cu in untreated and amended soils are shown in Figs. 2, 3, and 4. The first two fractions, namely the exchangeable fraction (F1) and the carbonate fraction (F2), are considerably more labile or mobilizable, and also more easily bioavailable, thereby posing greater risks to the environment and human health (Tessier et al. 1979; Lee et al. 2011b). Initially, Cd existed in a more available/mobile form than the other two metals, with 40 % of total Cd in the exchangeable and carbonate-bound fractions (F1–F2), whereas Pb and Cu in the non-amended soil were mainly in the residual fraction (F5) (64.2 and 52.0 % of total content, respectively), with relative levels in the low mobile fractions (F1–F2, 6.8 and 4.9 % of total content, respectively). After the amendments, heavy metals were found less in the mobile/available (F1–F2) fractions and more in the stable/residual (F3–F5) fractions.

Fig. 2
figure 2

Distribution of Cd fractions in different amendments addition soils

Full size image
Fig. 3
figure 3

Distribution of Pb fractions in different amendments addition soils

Full size image
Fig. 4
figure 4

Distribution of Cu fractions in different amendments addition soils

Full size image

The addition of the three amendments decreased the F1 fraction of Cd markedly with the largest reductions (13.6, 31.9, and 29.6 %) being reached in the 4 ‰ MP, MF, and MFP treatments, respectively (Fig. 2). The level in the F1 was mainly shifted to F3 fraction which increased 18.9, 15.0, and 26.9 % in 4 ‰ MP, MF, and MFP treatments, respectively. However, the F5 fraction of Cd decreased 13.5–24.8 % and 20.8–28.0 % in MP and MFP treatments.

Figure 3 shows that the amendments mainly shifted Pb from the F1 fraction to the F3 and F4 fractions. Significant changes were observed in the 4 ‰ MP, MF, and MFP treatments, where the exchangeable (F1) fraction decreased 56.4, 42.5, and 52.4 %, respectively, while the Fe-Mn oxide (F3) fraction increased 8.2, 5.7 and 10.2 %, respectively, and the organic combined (F4) fraction increased 9.8, 8.5, and 9.3 %, respectively.

As in the case of Cd and Pb, the amendments resulted in a shift of Cu from the labile fractions (F1 and F2) to more stable fractions (F3–F4) (Fig. 4). In the 4 % mixtures, the F1 fractions of Cu decreased 29.1 % (MP), 29.8 % (MF), and 34.3 % (MFP), while at the same time, the F3 fraction and/or F4 fractions increased markedly.

The mobility and/or availability of metals in soil can also be assessed on the basis of relative fractions weakly bound to soil components (i.e., F1 + F2). A relative index of metal mobility was calculated as “the mobility factor” (Narwal et al. 1999) using the following equation:

$$ \mathrm{M}=\frac{\left(F1+F2\right)}{\left(F1+F2+\mathrm{F}3+\mathrm{F}4+\mathrm{F}5\right)}\times 100 $$

where “M” is the relative content of the mobile fraction (F1 + F2). As can be seen in Table 4, initially, the mobility factor was 40.3, 6.93, and 4.93 % for Cd, Pb, and Cu, respectively. The amendments decreased the mobility factor of heavy metals significantly (P < 0.05), with the lowest M value of Cd, Pb, and Cu being observed in 4 ‰ MF-amended soil, MP-amended soil, and MF-amended soil, respectively (Table 4).

Table 4 Mobility factor of heavy metals in different amendments addition soils
Full size table

In this study, the application of amendments accelerated the reduction of the labile heavy metal pool (Figs. 2–4, Table 4). Increase of soil pH in these treatments seems to be the main reason for the reduction. A significant negative correlation was got between exchangeable fraction of heavy metals and soil pH values for all the treatments (r = −0.893**, −0.965**, and −0.961** for Cd, Pb, and Cu, respectively), indicating that the increase of soil pH duo to the addition of amendments led to the decrease in solubility of Cd, Pb, and Cu. Similar results have been reported in other studies (Janoš et al. 2010; Padmavathiamma and Li 2010; Ahmad et al. 2012).

Soil pH regulates the ionization of pH-dependent exchange sites on organic matter and oxide clay minerals, and the increased pH improved negative surface charge of soil and ionized pH-dependent exchange sites, raising cation-exchange capacity (CEC) and gets much more metals adsorbed on soil particles (Mench et al. 2000; Appel and Ma 2002; Bolan et al. 2003; Hong et al. 2007). What’s more, soil pH controls dissolution/precipitation reactions and influences metal speciation in soils (Adriano et al. 2004). Therefore, the increased soil pH converted heavy metal to much more stable fractions.

In addition to increased soil pH, there are other factors that affect the immobilization of Cd in the soil. In MP treatments, the increase in available P of soil may play a role in the retention of Cd, where a negative correlation between available P and exchangeable Cd was found (r = −0.468). Cd can be stabilized by co-adsorption of phosphate as an ion pair and surface complex formation of Cd on the phosphate compound in phosphoric amendment remediated soil (Bolan et al. 2003). Typically, stable and insoluble Cd phosphates may be formed due to the elevated available P content in MP-treated soil (Matusik et al. 2008). All three amendments raised the level of available Si in the soil significantly. In addition, a significant negative correlation between available Si and exchangeable Cd was observed; specifically, r = −0.886*, −0.956*, and −0.957* for MP, MF, and MFP, respectively. Moreover, much more Cd was found in Fe-Mn oxide-bound fraction in the amended soil compared to non-amended soil (Fig. 2). These results suggest an interaction between Si and Cd, and in fact, previous research has found that Si induced co-precipitation of Cd and Si in soil is a mechanism responsible for the reduction of labile Cd in amended soil (Sarwar et al. 2010).

Besides the increase of soil pH, the retention of Pb in MP treatments could be attributed to the P content of MP, as well. Phosphorus in soil can form stable and low-solubility phosphates with Pb (McGowen et al. 2001) and even generate much more stable substances, such as pyromorphite (Cao et al. 2002; Chen et al. 1997). Successful immobilization of Pb in soil requires sufficient phosphorus to provide free phosphate ions in the soil (Ma et al. 1995). By the addition of phosphate amendment (MP), there was an increase of 11.68–50.65 % available P in the soil (Table 3), thereby facilitating the formation of stable lead phosphates and decreasing the labile fractions of Pb. The significantly negative relationship between labile fraction Pb and available P in MP-amended soil (r = −0.832*) confirmed this possibility indirectly. Pb can also be stabilized by being adsorbed on the surface of the phosphate rock particles and fixed by subsequent cation exchange between phosphate rock Ca2+ and Pb2+ in soil solution (Shashkova et al. 1999). In MF and MFP amended soil, available Si and Ca negatively correlated with exchangeable fraction and the sum of exchangeable nd carbonate bound fractions (r = −0.934** and −0.759*, for F1 fraction and r = −0.963** and −0.886* for F1 + F2, respectively), indicating an interaction between Pb, Ca, and Si. Similar studies have reported that Pb may be immobilized by being entrapped into calcium-silicate-hydrate in amended soil (Badreddine et al. 2004; Cao et al. 2008; Ahmad et al. 2012).

Although the increase of soil pH was the main reason for the reduction of exchangeable Cu (Kumpiene et al. 2008), the decrease of labile Cu can also be ascribed to the precipitation of Cu–phosphate minerals (Garrido et al. 2005). A significant negative correlation between exchangeable Cu and available P in MP-amended soil (r = −0.947*) supports this possibility indirectly; furthermore, a similar result was reported by Garrido et al. (2005). In addition, the retention of Cu in this study may reflect the precipitation of Cu oxy-hydroxides and formation of complexes on the surface of Al oxy-hydroxides (Garrido et al. 2005).

3.4 Effect of the Amendments on the Phytoavailability of Heavy Metals

The application MP, MF, and MFP significantly decreased the concentrations of heavy metals in the shoot of pakchoi, except for the 0.5 ‰ MFP treatment for shoot Pb (Fig. 5). Compared with non-amended treatments, all the amended treatments decreased the Cd, Pb, and Cu in the shoot of pakchoi by 10.83–43.92 %, 2.87–47.5 %, and 21.14–57.07 % of, respectively. The lowest Cd, Pb, and Cu concentrations in pakchoi shoot were observed in 2 ‰ MF-amended treatment, 4 ‰ MFP-amended treatment, and 4 ‰ MF-amended treatment, respectively.

Fig. 5
figure 5

Concentrations of Cd, Pb, and Cu in the shoot of pakchoi in different treatments (different letters identify the significant difference (P < 0.05) among treatments)

Full size image

Similar to the reduction of heavy metals in the shoot, the concentrations of heavy metals in the plant root were significantly reduced by the addition of amendments (Fig. 6). Compared with the control group, the addition of amendments reduced the Cd, Pb, and Cu in the root of pakchoi by 2.35–39.3 %, 46.14–84.51 %, and 32.63–79.26 %, respectively. The largest reduction of Cd was observed in 2 ‰ MF-amended soil, and the lowest Pb and Cu were both observed in 4 ‰ MFP-amended soil.

Fig. 6
figure 6

Concentrations of Cd, Pb, and Cu in the root of pakchoi in different treatments (different letters identify the significant difference (P < 0.05) among treatments)

Full size image

The decreased phytoavailability of heavy metals can be attributed to the redistribution of metal species (from mobile and/or available fractions to stable and/or unavailable fractions) by the addition of amendments (Figs. 2, 3, and 4 and Table 4), which reduced heavy metal uptake by pakchoi directly. The significantly positive correlation between the concentrations of heavy metals in the F1 fraction and the heavy metals the plant absorbed indicate this possibility indirectly (r = 0.369/0.689**, 0.629*/0.513, and 0. 851**/0.872** for shoot/root Cd, Pb, and Cu, respectively). Lee et al. (2011a) also found that the correlation coefficients were significantly higher between shoot metal concentrations and extractable metal contents than between shoot metal concentrations and other parameters.

In addition to the reduction of bioavailable heavy metals in soil, the significant increase of available Ca (Table 3) can be related to the decrease of Cd uptake by plant. Specifically, it has been found that Ca as a plant nutrient can help to alleviate Cd toxicity where both Ca and Cd compete for the same Ca channels in plants (Nelson 1986; Tlustos et al. 2006). The application of MF and MFP was more effective in suppressing Cd uptake by the plants, probably due to its higher content of available silicon (Si), results that are similar to those reported by Chen et al. (2000).

Similar to the effect of Si on Cd uptake, Pb could be retained on the surface of the root due to binding to ion exchange sites and extracellular precipitation, mainly in the form of Pb phosphates especially in the MP treatment (Jarvis and Leung 2002).

3.5 Effect of the Amendments on Plant Growth

The effect of the amendments on the growth of pakchoi is shown in Fig. 7, which shows that almost all the amendments promoted the growth of pakchoi and increased its shoot biomass, root biomass, and height by 73.3–884 %, 53.2–406 %, and 8.64–146 %, respectively, relative to the control. Only in the 0.5 ‰ MFP treatment was there a slight decrease in shoot biomass and root biomass observed compared to the control.

Fig. 7
figure 7

The height and biomass of pakchoi in different treatments

Full size image

Correlations were computed for the pakchoi growth vs various parameters in the soils amended with MP, MFP, and MFP (Table 5). A significantly positive correlation between the growth of pakchoi (biomass and height) and soil pH was observed for soils amended (Table 5). Therefore, the elevated pH was more suitable for the growth of pakchoi. This beneficial effect of soil pH neutralization on the biomass of pakchoi has also been observed in other studies (Zhou et al. 2005; He et al. 2009). The increased mineral nutrients (K, Ca, Mg, Si) also showed positive correlations with pakchoi growth (Table 5), especially the significantly elevated available Ca, Mg, and Si. Conversely, the significantly negative correlation between the growth of pakchoi (biomass and height) and the concentrations of heavy metals in the labile fraction (Table 5) suggests that these soil amendments can improve the growth of pakchoi by decreasing the availability of heavy metals in soil pools. Moreover, the correlation between heavy metals in pakchoi and the growth of pakchoi is significantly negative, indicating that the decrease of heavy metal uptake by the plant reduced the damage to pakchoi, a result similar to that observed by Chen and Zhang (Chen et al. 2011; Zhang et al. 2013). Overall, these results suggested that the increased pH, the elevated mineral nutrients, and the decreased toxicity of heavy metals associated with the soil amendments created much more suitable soil conditions for pakchoi growth.

Table 5 Correlation coefficients between soil properties studied and pakchoi growth index
Full size table

4 Conclusion

This study evaluated the application of three modified industrial by-products as soil amendments in the restoration of heavy metal-contaminated agricultural lands. Cd, Pb, and Cu immobilization, soil quality improvement, and pakchoi growth response were evaluated when these soil amendments were applied at four application rates. The most effective mixture for all the amendments was 4 ‰ w/w.

The application of the three modified industrial by-products was effective in immobilizing Cd, Cu, and Pb in soil and reducing their bioavailability to pakchoi. The significant conversion of heavy metal fractions (from labile fractions to stable fractions) was mainly attributed to the increased soil pH and the function of elevated Ca, Si, and P contents. MF and MFP were more effective in the immobilization of Cd, mainly due to elevated levels of available Si in soil; the P-rich amendment (MP) was more effective for Pb retention; and all three types of amendments were effective in decreasing the availability of Cu.

In addition to the immobilization of heavy metals, the amendments improved soil quality. The addition of the three alkaline amendments neutralized the acidity of the soil and increased soil pH by 0.13–1.26 units. Since all of the amendments are rich in available Ca, Mg, and Si, especially MF and MFP, the addition of the amendments increased soil Ca, Mg, and Si content as a whole. The application of P-rich amendment (MP) enhanced soil available P significantly (p < 0.05), and the increase of available K reached statistical significance in MF and MFP treatments.

All the amendments promoted the growth of pakchoi and increased the shoot biomass, root biomass, and height of pakchoi. The increased pH, the elevated mineral nutrients, and the decreased toxicity of heavy metals created a much more suitable soil condition for pakchoi growth due to the amendments addition.

We conclude that the application of the three modified industrial by-products as soil amendments had a demonstrably positive impact on the remediation and revitalization of heavy metal-contaminated agricultural soil. However, further studies are needed in order to investigate the effects of the amendments on the biological environment of heavy metal-contaminated soil.