Multidimensional Effects of Revegetation on Antimony Mine Waste Slag: From Geochemical Responses to Ecological Risk Regulation

Abstract

Revegetation is considered a sustainable option for mine area remediation. However, the sustainability and risk evolution of revegetation for large antimony mine slag remain incompletely understood. In this study, we focused on the revegetation project of the waste slag heap of XKS, the world’s largest antimony mine. Systematically analyzed the physicochemical properties, total metal(loid) content, and BCR sequential extraction and applied the modified comprehensive pollution risk assessment (MCR) method to evaluate ecological risk evolution. The results showed that revegetation can effectively increase the nutrient content, and the total content of nitrogen and phosphorus maximally increased by 5.15 and 1.89 times, respectively, after 10 years of remediation. Long-term revegetation could mitigate the metal(loid) contamination, and the average contents of As and Sb decreased by 88.72–93.18% and 93.47–89.87%, respectively. BCR analysis showed that the percentage of residual As and residual Sb increased from 64.75% and 85.88% to 78.38% and 91.58%, respectively. The MCR assessment method showed that revegetation could effectively reduce the ecological risk level. This study provides important multidimensional evidence for the ecological restoration of antimony mining areas, which can provide practical guidance for subsequent slag management and risk control.

1. Introduction

China has the largest antimony reserves in the world and is also a globally important producer (accounting for about 78% of the total global production capacity) [1]. Although antimony exploitation has brought enormous benefits to social and economic development, it has also affected the environment of the mining area, especially in the area where the mining solid wastes (such as smelt slag and tailings) are deposited [2,3]. These large amounts of solid waste are generated during the extraction, beneficiation, and smelting of metal ores [4], which are mostly simply disposed of by open-air stockpiling. Under the effects of weathering, erosion, rainfall leaching, and speciation transformation, the residual metal(loid)s may be released and migrate, which consequently causes a series of ecological problems such as acidic mine drainage, contamination of the surrounding soil and water, ecological degradation, etc. [3,5]. Taking the Qinglong Dachang mine, a typical antimony mining area, as an example, the concentrations of antimony (Sb) and arsenic (As) in the tailings reach as high as 13,896 mg/kg and 657 mg/kg; in the smelting waste slag, Sb and As concentrations peak at 23,559 mg/kg and 1177 mg/kg, while mining-contaminated soils contain Sb and As at 2885 mg/kg and 63.8 mg/kg, all values significantly higher than the background levels [6,7]. Similarly, Sb concentrations ranging from 2735 to 4517 mg/kg were found in the contaminated soil surrounding the Hillgrove mine waste in Australia, which posed a critical threat to regional ecological security [8]. Therefore, proper management of mining solid wastes and implementation of measures to minimize the release of metal(loid)s are essential [9,10].

Traditional mine remediation techniques, such as chemical stabilization [11,12,13], soil washing [14], ex situ stabilization [15], and electrokinetic remediation [16], can effectively solidify or remove the metal(loid)s in a short time, but they also have the problems of high remediation cost and possible risk of secondary pollution (the metal(loid)s may be released again when the solidifying agent fails). In contrast, revegetation is regarded as a more sustainable solution for mine site remediation due to its small side effects, low cost, and simultaneous realization of contaminant migration control and ecological function recovery [17,18].

In recent years, revegetation technology has achieved excellent effects in the ecological treatment of mining areas, including the increase in soil nutrients, the improvement in soil structure, the decrease in metal(loid)s concentration, and the restoration of microbial communities. For example, Lin et al. [19] chose six species of herbaceous plants (Melilotus officinalis, Xanthium sibiricum, Festuca elata, Zoysia japonica, Amaranthus tricolor L., and Artemisia desertorum) to carry out potting experiments for revegetation restoration of gold mine tailings, and the results showed that revegetation treatment could effectively improve the physicochemical properties, and the salinity of the tailings decreased significantly while the nutrients increased gradually. Wan et al. [20] found that the content of sodium (Na), aluminum (Al), and chromium (Cr) elements in bauxite slag leachate was significantly reduced after revegetation treatment over a 100-day microcosm experiment. Similarly, Liu et al. [21] conducted a field experiment in a gold tailings pond and showed that the revegetation could effectively reduce the metal(loid) content in the rhizosphere and enhance the nutrients. Beyond short-term revegetation studies, several researchers have focused on the effects of revegetation on the chemical behavior of metal(loid)s. For example, Burke et al. [22] analyzed depth profile samples from Pb-Zn tailings at different revegetation times (0–8 years) and showed that revegetation can drive the transformation of Sb and As from the reduced state (Sb³⁺ and As³⁺) to the oxidized state (Sb⁵⁺ and As⁵⁺), which provides valuable insights into understanding the geochemical behavior of the metal(loid)s under revegetation. However, the systematic effects of revegetation in large-scale antimony mining areas are still lacking, particularly regarding comprehensive monitoring of metal(loid) migration behavior, bioavailability changes, and ecological risk evolution. In addition, existing ecological risk assessment methods for mining areas, such as the geo-accumulation index method and the potential ecological risk assessment method, all focus on the total content of metal(loid)s [9,23]. It should be noted that only the bioavailable fraction of metal(loid)s has a strong mobility and can be easily absorbed by organisms. Therefore, only using the total content as an indicator may exaggerate the actual risk level, which may lead to an unnecessary increase in the management cost [24,25,26]. Therefore, it is critically needed to conduct the multidimensional monitoring and comprehensive assessment of the revegetation effects in antimony mining areas, which will fill existing research gaps and inform environmental management strategies for these regions.

Thus, the XKS antimony mine in Hunan Province, which is the largest antimony mine in the world and known as the “antimony capital” [27], was selected as a typical research area for this study. Through a systematic comparison of geochemical parameters, total metal(loid) content, speciation distribution, and the comprehensive risk between the unremediated waste piles and the revegetated restored area. This study aimed to (1) reveal the effects of revegetation on the geochemical parameters of the antimony mine area, (2) investigate whether revegetation could stably and persistently reduce the total content of metal(loid)s and change their chemical patterns, and (3) assess whether the comprehensive risk level of the antimony mine waste dump area has been reduced to a controllable level after the revegetation. The results of this study provide a theoretical basis for the scientific evaluation of the ecological effect of antimony mine revegetation and guidance for the subsequent management of mine waste and ecological risk control.

2. Materials and Methods

2.1. Study Area and Sample Collection

This study was carried out in the XKS mining area (111°25′ E–111°31′ E, 27°43′ N–27°49′ N), located in the central part of Hunan Province, China (Figure 1), which has the largest antimony reserves and production in the world. It has been officially mined for more than 120 years since 1897 [28] and has a total area of about 26 km². The area has a subtropical continental monsoon climate with an average annual rainfall of 1354 mm and an average annual temperature of about 16.7 °C [2,29]. Intensive mining activities have led to the accumulation of a large amount of solid waste, and the total cumulative amount of waste slag generated in the mining area reaches more than 7500 × 10⁴ t, with about 350 slag heaps, mainly including smelting slag and tailings, etc. [30]. Recently, the local government has greatly improved the ecological environment of the mining area through the program of ecological remediation of a large number of historical slags, and the vegetation coverage of the ecological remediation area has been significantly increased.

Based on the time series of the slag ecological remediation project, we selected the unremediated smelter slag (marked as US), unremediated tailings (marked as UT), the smelter slag heap after 5 years of remediation (marked as RS5), the smelter slag heap after 10 years of remediation (marked as RS10), and the tailings piles after 10 years of remediation (labeled as RT10), respectively, for sample collection. In addition, native vegetation areas (marked BS) were also selected for sampling to represent pre-disturbance levels.

The ecological remediation of the historical smelter slag heap and tailings piles followed standardized protocols implemented by the local government. For all remediation areas included in this study, the same core procedures were applied. Revegetation primarily utilized native and locally adapted plant species selected for their tolerance to the challenging substrate conditions and potential for stabilizing contaminants. In order to maintain consistency, the dominant species across all remediated sampling sites was Miscanthus sp., a perennial grass known for its robust growth, extensive deep root system, and reported tolerance to metal(loid)s (can accumulate a great amount of metal, such as the average concentrations of As and Sb in plant roots reaching 38.5 and 177.3 mg/kg, respectively) [31,32,33]. Prior to planting, the substrate surface was mechanically leveled and loosened. Plant establishment relied solely on natural precipitation; no supplemental irrigation was applied after the revegetation process. Miscanthus sp. were established primarily through seeding.

In July 2024, five composite samples were collected from each of the study areas using a mixed multipoint sampling method, with three replicate samples collected from each sample site and mixed into one composite sample using the quadratic method. A total of 30 composite samples were collected; all samples were taken from near-surface depths (0–20 cm), and the collected samples were placed in sample bags and immediately transported back to the laboratory. After natural drying, milling, and sieving, they were used for subsequent geochemical characterization, total metal(loid)s, and metal(loid)s speciation analyses. All chemicals and reagents used in this study were at least of analytical grade.

2.2. Physicochemical and Geochemical Characterization

The pH value was measured by a pH meter (PE28-Standard, Mettler Toledo, Shanghai, China) at a solid–liquid ratio of 1:2.5. Organic carbon content (Corg) in the samples was analyzed by potassium dichromate oxidation [34,35]. Total nitrogen content (TN) was determined using an elemental analyzer (EA300, EuroVector, Milano, Italy). Total phosphorus (TP) content was determined by the molybdenum blue colorimetric method (UV3200, Shimadzu, Japan). Aluminum oxide (Al₂O₃), calcium oxide (CaO), ferric oxide (Fe₂O₃), silicon dioxide (SiO₂), magnesium oxide (MgO), sodium oxide (Na₂O), and potassium oxide (K₂O) contents of the samples were determined by X-ray fluorescence spectrometer (XRF, Axios Max, PANalytical, Almelo, The Netherlands).

2.3. Analysis of the Total Metal(loid) Content

The content of Cd, Ni, Pb, Zn, Cu, and Cr was determined by inductively coupled plasma atomic emission spectrometry (IRIS Intrepid, Thermo Electron, Waltham, MA, USA) after microwave digestion (MARS Xpress, CEM, Stallings, NC, USA) by mixed acid digestion method, according to the method of inductively coupled plasma emission spectrometry for the determination of metal elements in solid wastes (HJ781-2016 [36]). The content of As, Hg, and Sb was determined by hydride generation-atomic fluorescence spectrometry (HG-AFS, Haikuang, China) after microwave digestion using the aqua regia (HNO₃: HCl = 1:3, V/V) digestion method. The detection limits of each analytical method met the requirements of Geochemical Evaluation of Land Quality (DZ/T0295-2016 [37]).

2.4. Metal(loid)s Chemical Speciation Analysis

The BCR sequential extraction method was used to analyze the chemical speciation of As and Sb elements in the samples, and the metal(loid) forms were classified into exchangeable (EXC), reducible (RED), oxidizable (OXI), and residual (RES) fractions. The detailed extraction process is described in the studies of Li [38]. The elemental contents of As and Sb in the extraction solution were determined by hydride generation-atomic fluorescence spectrometry (HG-AFS, Haiguang, Beijing, China).

2.5. Modified Comprehensive Pollution Risk (MCR) Method

The MCR method used in this study is based on Jiang and Li et al. [25,39]. The method assumes seven pollution levels (

Table 1. Degree of contamination and environmental risk assessment criteria for As and Sb.

Category	RAC	Igeo	EF	ER	CRS	Degree of Pollution
1	<1%	≤0	-	-	6	None
2	1–10%	0–1	<2	<40	6–8	Low
3	10–30%	1–2	2–5	40–80	8–12	Moderate
4	-	2–3	5–20	80–160	12–15	Considerable
5	30–50%	3–4	20–40	160–320	15–20	High
6	>50%	4–5	>40	>320	20–24	Very high
7	-	5–10	-	-	25	Extremely serious

RAC: risk assessment code; I_geo: geo-accumulation index; EF: enrichment index; ER: ecological risk; CRS: comprehensive pollution risk score.

) and assigns the scores of no pollution, low pollution, moderate pollution, considerable pollution, high pollution, very high pollution, and extremely serious pollution as 0, 1, 2, 3, 4, 5, and 6, respectively. First, the risk assessment code (RAC) evaluation results were calculated using Equation (1), the geo-accumulation index (I_geo) was calculated using Equation (2), the enrichment index (EF) was calculated using Equation (3), and the potential ecological risk factors (ER) were calculated using Equation (4). Subsequently, the above four evaluation results were assigned scores according to pollution levels with reference to the evaluation criteria of RAC, I_geo, EF, and ER in Table 1. Then Equation (5) was used to calculate the comprehensive pollution risk score (CRS) for metal(loid)s. Finally, the comprehensive evaluation results were obtained by referring to the CRS comprehensive evaluation criteria in Table 1.

$$RAC={\displaystyle \frac{C_{EXC}}{C_{tot}}} \times 100\%$$

(1)

where C_EXC is the exchangeable fractions content of element i and C_tot is the total content of element i.

$$I_{geo}={log}_2{\displaystyle \frac{C_i}{1.5B{E}_i}}$$

(2)

where C_i is the total content of element i and B_Ei is the average content of element i in the collected background soil.

$$EF={\displaystyle \frac{{\left(C_i/C_n\right)}_{sample}}{{\left(C_i/C_n\right)}_{baseline}}}$$

(3)

where C_n is the total content of the Fe, C_i is the total content of element i, sample is the collected sample, and baseline is the collected background soil.

$$ER=T_r^i\times \left(C_s^i/C_n^i\right)$$

(4)

where Tⁱ_r is the toxicity factor of element i, in which the toxicity factors of As and Cd are 10 and 30, respectively [25], Cⁱ_s is the total content of element i, and Cⁱ_n is the total content of element i in the background soil.

$$CR{S}_i=S_i\left(RAC\right)+S_i\left(I_{geo}\right)+S_i\left(EF\right)+S_i\left(ER\right)$$

(5)

where CRS_i is the composite pollution risk score for element i, S_i(RAC) is the RAC pollution level score for element i, S_i(I_geo) is the I_geo pollution level score for element i, S_i(EF) is the EF pollution level score for element i, and S_i(ER) is the ER pollution level score for element i.

2.6. Statistical Analysis

Except where noted, data in this study are presented as mean ± standard deviation. Statistical analysis was performed using SPSS 26.0 software (IBM Corporation, Armonk, NY, USA), and Spearman correlation analysis was used to determine the relationship between the comprehensive ecological risk of metal(loid)s and chemical characteristics, total content, and chemical speciation. Mapping was performed using Origin 2024 software (OriginLab, Northampton, MA, USA).

3. Results and Discussion

3.1. Effects of Revegetation on the Physicochemical and Geochemical Properties Characteristics of Mining Areas

The physicochemical and geochemical characteristics of the original smelting slag and tailings, as well as the smelting slag heap and tailing piles after revegetation treatment, were analyzed, including the basic physicochemical properties such as pH, TN, TP, and OM, and the chemical compositions such as Fe₂O₃, Al₂O₃, and SiO₂.

As shown in

Table 2. Changes in physicochemical and geochemical characteristics after revegetation.

	pH	TN (g/kg)	TP (g/kg)	Corg(%)	Al2O3(%)	CaO(%)	Fe2O3(%)	SiO2(%)	MgO(%)	Na2O(%)	K2O(%)
BS	5.51 ± 0.03	1.21 ± 0.10	0.415 ± 0.04	1.53 ± 0.38	14.86 ± 1.24	0.10 ± 0.03	5.42 ± 0.63	66.50 ± 2.69	0.44 ± 0.02	0.21 ± 0.03	1.33 ± 0.11
US	8.88 ± 1.71	0.38 ± 0.01	0.177 ± 0.03	1.94 ± 0.51	11.86 ± 2.05	2.98 ± 1.51	4.28 ± 0.25	65.93 ± 1.37	0.61 ± 0.12	0.67 ± 0.08	1.30 ± 0.28
RS5	6.68 ± 1.33	0.76 ± 0.14	0.243 ± 0.01	0.24 ± 0.06	8.47 ± 0.59	0.28 ± 0.10	3.64 ± 0.23	81.23 ± 0.84	0.32 ± 0.05	0.08 ± 0.01	0.92 ± 0.03
RS10	5.35 ± 1.27	0.95 ± 0.01	0.255 ± 0.02	1.08 ± 0.09	8.61 ± 0.30	0.21 ± 0.13	3.27 ± 0.16	78.70 ± 0.28	0.55 ± 0.01	0.17 ± 0.01	1.13 ± 0.02
UT	7.93 ± 0.63	0.21 ± 0.01	0.19 ± 0.04	0.42 ± 0.02	4.71 ± 0.26	4.56 ± 0.46	1.38 ± 0.21	82.2 ± 0.34	0.31 ± 0.04	0.06 ± 0.01	0.61 ± 0.04
RT10	7.71 ± 0.25	1.07 ± 0.13	0.355 ± 0.02	1.04 ± 0.15	14.57 ± 2.74	1.23 ± 0.60	5.96 ± 1.12	65.6 ± 7.92	0.82 ± 0.13	0.13 ± 0.01	1.58 ± 0.01

US = unremediated smelting slag; RS5 = smelting slag heap remediated after 5 years; RS10 = smelting slag heap remediated after 10 years; UT = unremediated tailings; RT10 = tailings piles remediated after 10 years; TN = total nitrogen; TP = total phosphorus; Corg = organic carbon.

, the untreated original smelting slag and tailings were both alkaline (pH 8.9 and 7.9, respectively). Consistent with the results of other studies [40], the pH values of both smelting slag heaps and tailings piles were significantly reduced after revegetation and showed a gradual decreasing trend with the increase in revegetation time; the pH values of RS5 and RS10 were 6.7 and 5.4, respectively. Related studies have found that plant root exudates, plant apoplasts, and their degradation products may contribute to pH reduction [40,41], in addition to H⁺ produced by nitrification reactions, and the release of H⁺ and organic acids promoted by plant rooting activities can also lead to a reduction in environmental pH [42].

The content of TN in the smelting slag heap after revegetation was significantly higher than that in the unremediated slag, and the content of TN increased gradually with the extension of the revegetation time. After 10 years of revegetation, the average TN contents in the RS10 and RT10 areas were 0.95 g/kg and 1.07 g/kg, respectively, which were 2.5 and 5.15 times higher than those in the US (0.38 g/kg) and UT (0.21 g/kg) areas and were close to the background soils (1.21 g/kg). As a dominant plant in the remediated area, Miscanthus sp. can accumulate phosphorus and nitrogen nutrients in the rhizosphere [43]. In addition, related studies reported that the decomposition of plant litter and organic matter during revegetation [40,44] and that carbon-sequestering microorganisms in the rhizosphere of plants may also increase the nitrogen content of the surrounding environment [45].

The content of TP also showed a gradual cumulative increase after revegetation, and after 10 years of remediation, the average content of TP in the RS10 and RT10 areas was 0.26 g/kg and 0.36 g/kg, respectively, which were 1.44 and 1.89 times higher compared with that in the US (0.18 g/kg) and UT (0.19 g/kg) and were gradually close to the background soils (0.42 g/kg). Similar to the results of this study, several other research on mining sites have also reported an increase in TP content after remediation [40,44,45], which could be attributed to the release of root exudates during revegetation that can accelerate the weathering process of minerals, such as phosphate, and thus release phosphorus [40,44].

The positive effects of revegetation on the recovery of soil carbon sequestration have been widely reported [46,47], and the OM content in the tailings area also showed a significant increase in this study, which is presumed to be attributed to the positive impacts of the production and decomposition of the plant litter [46]. However, the unremediated smelting slag showed abnormal OM content, which was also found in the previous research [45], and this might be attributed to the presence of reduced substances in the slag, which led to the artifact of higher organic carbon content [45]. The results of chemical composition analysis showed that the oxides were mainly SiO₂, Al₂O₃, Fe₂O₃, MnO, MgO, CaO, Na₂O, and K₂O, and there was no significant difference in the oxides except for SiO₂, CaO, and MgO.

In general, the revegetation treatment resulted in a better improvement in environmental pH, nutrient elements, and organic matter content, and the extension of the revegetation time had a positive effect on the improvement in physicochemical properties.

3.2. Effect of Revegetation on Metal(loid) Content in Mining Areas

The contents of metal(loid)s in smelting slag and tailings after revegetation are shown in Figure 2. The contents of As, Cd, Hg, Cu, Ni, Pb, Sb, and Zn in the remediated smelting slag heap and tailings piles were significantly lower than those in the unremediated US and UT. Compared to the unremediated US, the average contents of As, Cd, Hg, Ni, Pb, Sb, Zn, and Cu in the RS5 area decreased by 75.09%, 91.12%, 49.17%, 62.15%, 88.89%, 63.15%, 69.76%, and 61.35%, respectively, and the average contents of As, Cd, Hg, Ni, Pb, Sb, Zn, and Cu in the RS10 area decreased by 93.18%, 92.00%, 73.39%, 53.85%, 89.50%, 89.87%, 62.99%, and 62.68%, respectively. Similarly, the average content of As, Cd, Hg, Ni, Pb, Sb, Zn, and Cu decreased by 88.72%, 60.45%, 83.30%, 22.71%, 23.26%, 93.47%, 23.04%, and 14.54%, respectively, in the RT10 area compared to the UT.

Results showed that revegetation could effectively control the metal(loid) pollution problem of smelting slag heaps and tailings piles, which exhibited a tendency of gradual reduction of metal(loid) content with the prolongation of revegetation time. Luo [48] researched the effect of revegetation on Mn, Zn, Pb, and Cd in manganese mine wasteland and also found that the metal content of the soil was significantly reduced after revegetation compared to the unremediated mine waste sites, and it gradually decreased with the increase in the time of revegetation, and after 9 years of revegetation, the metal content was 85.29% lower than that of the unremediated plots. Sun’s research [40] also found that after 7 years of revegetation, As and Sb concentrations in Pb-Zn smelting slags were significantly lower than in unremediated control slags. This may be due to the influence of plant root exudates and microbial activities on the chemical formation and adsorption-resolution behavior of metal(loid)s in the waste slag, which promotes the transport and transformation of metal(loid)s [49]. Related studies have reported that root exudates of Miscanthus can promote metal uptake by releasing organic acids, chelating agents, and other compounds [31]. Xue et al. [32,50] found that the enrichment concentration of As and Sb in the root system of Miscanthus could reach 38.53 and 177.29 mg/kg. Similarly, Xiao et al. [43] found that the shoot of Miscanthus contained 63.41, 360.53, and 158.5 mg/kg of Cu, Zn, and Pb, respectively, and the Pb and Zn content in the roots reached 1449.09 and 1114.33 mg/kg, respectively. The results of the above-related studies all indicated that Miscanthus has the potential to absorb metal(loid)s. In addition, with the increase in the vegetation restoration time, the rhizosphere organic matter and nutrient elements accumulated continuously, which effectively improved the soil quality, and the vegetation growth gradually increased; thus, the ability of absorption and accumulation of metal(loid)s also increased accordingly [31,48].

Furthermore, compared with the background soil around the mine site, the content of metal(loid) elements after ten years of revegetation was almost close to the background soil except for As, Cd, Hg, and Sb. The content of As, Cd, Hg, and Sb in the RS10 area was 60.80, 0.74, 2.13, and 604.50 mg/kg, respectively, and 37, 1.69, 0.62, and 128 mg/kg in the RT10 area, respectively. Compared with the screening values for Class II land use specified in the Soil Environmental Quality Risk Control Standard for Soil Contamination of Construction Land (Trial) (GB 36600-2018 [51]), except for localized residual Sb in RS10, the content of metal(loid) elements was below the risk control screening value.

Overall, revegetation is a kind of effective technology to control metal(loid) pollution, and with the increase in the period of vegetation restoration, the better the control effect of metal(loid) content in the mining area. After ten years of remediation, the total metal(loid) content of smelting slag heaps and tailings piles has decreased significantly.

3.3. Effect of Revegetation on the Chemical Speciation of Metal(loid)s in Mining Areas

The environmental behavior of metal(loid)s not only depends on their total concentrations but is also significantly influenced by their chemical speciation [9,40]. Based on the analysis results of the total concentration of metal(loid)s, As and Sb, which are relatively high in the total content of the samples, were selected as the key elements for further analysis of the metal(loid)s chemical speciation.

The results of the chemical speciation of As and Sb are shown in Figure 3. In the US, the percentage of the speciation of As was in the order of residual As (63.76%) > reducible As (28.36%) > exchangeable As (3.66%) > oxidizable As (3.23%). Compared with the UT, the percentage of reducible As was relatively higher, which was presumed to be mainly related to the potential presence of As in amorphous glassy phases and microstructural features (pores and defects) formed during high-temperature smelting, making As more susceptible to release under the reducing conditions of the BCR. After revegetation, the percentage of the speciation of As was in the order of residual As (73.70–78.97%) > reducible As (17.99–22.03%) > oxidizable As (2.55–3.96%) > exchangeable As (0.31–0.48%). Compared with the US and UT, the percentage of residual As gradually increased after revegetation, while the percentages of both reducible and exchangeable As decreased to different degrees.

The chemical speciation of Sb in the US was in the order of residual Sb (85.88%) > oxidizable Sb (12.63%) > reducible Sb (1.03%) > exchangeable Sb (0.46%), and after revegetation, the speciation of Sb changed to residual Sb (91.58–96.61%) > reducible Sb (1.76–4.84%) > oxidizable Sb (1.34–3.19%) > exchangeable Sb (0.29–0.38%). Both As and Sb in all samples were mainly in the residual fraction, indicating that most of the As and Sb are present in the lattice of primary and secondary minerals, which have low mobility and can be stable for a long period of time under the natural condition [27,40]. Compared with other speciation forms, exchangeable As (0.31–0.48%) and exchangeable Sb (0.29–0.76%) were the least fraction after revegetation, whereas the percentage of As and Sb in the reducible fraction was higher than that in the oxidizable fraction and exchangeable fraction, which was mainly attributed to the strong adsorption of As and Sb with the Fe-Mn oxides [27]. Under the influence of plant growth activities, exchangeable As and Sb can be transported by plants, while oxidizable and reducible As and Sb may be transformed by plant activation or by weathering and leaching [40]. In addition, increased weathering and other activities may lead to the combination of As and Sb with amorphous secondary minerals, thus causing an increase in the content of residual fractions [27,40,52].

In general, after revegetation, the percentage of exchangeable As and Sb in the slag, which are easy to migrate and transform, decreased significantly, while the percentage of residue formation increased gradually, indicating that the bioavailability of As and Sb in the slag decreased significantly with the revegetation. From the perspective of environmental risk, there is no doubt that this is beneficial for the control of metal(loid)s migration in the mining area.

3.4. Impact of Revegetation on Environmental Risks in Mining Areas

The environmental risks of As and Sb in mining areas after revegetation were evaluated by the Modified Comprehensive Pollution Risk Assessment method (MCR) [25,39]. The CRS scores of As pollution in each area before and after revegetation (Figure 4a) were US (17) > UT (15) > RS5 (11) > RS10 (7) > RT10 (6), and based on the CRS assessment criteria (Table 2), the values of the CRS_As in the unremediated US and UT areas were all higher than or equal to 15, which indicated that the unremediated US and UT had a high risk of As pollution. With the revegetation of the mine area, the values of CRS_As decreased significantly, and the values of CRS_As in the RS5, RS10, and RT10 areas were reduced by 6, 10, and 11 scores, respectively, compared with the US areas. Compared to the CRS assessment criteria (Table 2), the As pollution problems in RS10 and RT10 have been reduced to low or no pollution levels after 10 years of revegetation treatment. The results showed that the revegetation treatment could effectively reduce the environmental risk of As contamination; with the extension of the revegetation time, the environmental risk of As contamination gradually decreased. Meanwhile, the CRS scores of Sb contamination in each area (Figure 4b) were in the following order: US (19) > UT (16) > RS5 (15) > RS10 (10) > RT10 (6), which showed a similar rule with the As contamination. Both unremediated US and UT had the high risk level of Sb contamination, while the CRS_Sb values decreased significantly with the revegetation. After 10 years of revegetation, the CRS_Sb of the RS10 and RT10 areas were 10 and 6, respectively, which indicated that the ecological risk of the accumulation of Sb in the mining area was gradually reduced with the revegetation. This was consistent with the changes in the total amount of metal(loid)s and the transformation of chemical speciation. The results of correlation analysis (Figure 5) also demonstrated that the environmental risks of As and Sb were significantly correlated with their total content and the percentage of exchangeable fraction. In addition, under the effect of revegetation, the geochemical principles were changed, and nutrient elements such as N and P were accumulated in the rhizosphere, which effectively improved the quality of the soil [31,48]. Under the influence of plant root exudates and microbial activities, etc., the content of metal(loid)s and the percentage of bioavailable chemical forms can be effectively reduced, thus reducing environmental risks [35,40,49].

Compared to other environmental risk assessment methods, the results of I_geo, EF, and ER exhibited similar risk trends. However, there were some differences among the assessment methods, for example, the results of I_geo, EF, and ER showed that Sb contamination in the US reached a very high risk level, while the results of RAC showed that it was only a low contamination risk level. Previous studies have also shown that I_geo, EF, and ER values may overestimate the bioavailability and toxicity risk of As and Sb [25]. This is because the I_geo, EF, and ER values are calculated using the total content of metal(loid)s and do not consider the mobility and bioavailability of metal(loid)s, whereas the RAC calculations are assessed based on the percentage of bioavailable chemical formation of metal(loid)s in the contaminated site. Since most of As and Sb were in the residual fraction and the percentage of their exchangeable fraction was less than 10%, therefore, based on the RAC assessment criteria, they were shown as non-risk or low-risk level. The MCR method effectively integrates the total amount of metal(loid)s and their chemical formation, which can provide a more complete and accurate reference for the evaluation of the potential environmental risk [25,39].

In general, the results of the environmental risk assessment showed that revegetation could effectively reduce the environmental risk caused by metal(loid) pollution in the mining area, and with the extension of the revegetation period, the environmental risk value decreased. Revegetation is an effective technique for controlling metal(loid) pollution in mining areas.

4. Conclusions

In this study, we focused on the revegetation of waste slag heaps at XKS, the world’s largest antimony mine, investigating the response of environmental geochemical behaviors and the evolution of comprehensive ecological risks. Results indicated that revegetation could positively improve the geochemical characteristics of the slag heap, and with the increase in the treatment time, the pH value gradually decreased from strongly alkaline to nearly neutral, while the TN, TP, and Corg content gradually increased. Compared with the unremediated slag heap, the total content of metal(loid)s such as As and Sb decreased significantly after revegetation and gradually approached the background level of the original vegetation area. BCR analysis showed that the exchangeable fraction of Sb and As decreased significantly after revegetation, while the residual fraction increased. Based on the MCR method, the ecological risk of the area was reduced from the very high risk level to the less risky level. Generally, the revegetation technology in antimony mining areas can effectively improve the environmental quality and reduce the risk level. This study provides an effective insight into the ecological restoration of antimony mines, based on the multidimensional perspective of “geochemistry—speciation transformation—risk evolution”, which can provide a basis for the precise control of the remediation process. Based on the key findings of this study, the research topics that need to be further explored are (1) decoding rhizosphere processes, integrating metagenomics and metabolomics to explore how root exudates regulate metal(loid) speciation, and identifying functional microbiomes and (2) establishing long-term resilience monitoring, tracking ecosystem stability, and establishing the dynamic MCR forecasting models.