Effects of dissolved organic matter derived from cow manure on heavy metal(loid)s and bacterial community dynamics in mercury-thallium mining waste slag

Organic amendments in aided phytostabilization of waste slag containing high levels of heavy metal (loid)s (HMs) are an important way to control the release of HMs in situ. However, the effects of dissolved organic matter (DOM) derived from organic amendments on HMs and microbial community dynamics in waste slag are still unclear. Here, the effect of DOM derived from organic amendments (cow manure) on the geochemical behaviour of HMs and the bacterial community dynamics in mercury (Hg)-thallium (Tl) mining waste slag were investigated. The results showed that the Hg-Tl mining waste slag without the addition of DOM continuously decreased the pH and increased the EC, Eh, SO42−, Hg, and Tl levels in the leachate with increasing incubation time. The addition of DOM significantly increased the pH, EC, SO42−, and arsenic (As) levels but decreased the Eh, Hg, and Tl levels. The addition of DOM significantly increased the diversity and richness of the bacterial community. The dominant bacterial phyla (Proteobacteria, Firmicutes, Acidobacteriota, Actinobacteriota, and Bacteroidota) and genera (Bacillus, Acinetobacter, Delftia, Sphingomonas, and Enterobacter) were changed in association with increases in DOM content and incubation time. The DOM components in the leachate were humic-like substances (C1 and C2), and the DOC content and maximum fluorescence intensity (FMax) values of C1 and C2 in the leachate decreased and first increased and then decreased with increasing incubation time. The correlations between HMs and DOM and the bacterial community showed that the geochemical behaviours of HMs in Hg-Tl mining waste slag were directly influenced by DOM-mediated properties and indirectly influenced by DOM regulation of bacterial community changes. Overall, these results indicated that DOM properties associated with bacterial community changes increased As mobilization but decreased Hg and Tl mobilization from Hg-Tl mining waste slag.

Abstract Organic amendments in aided phytostabilization of waste slag containing high levels of heavy metal (loid)s (HMs) are an important way to control the release of HMs in situ. However, the effects of dissolved organic matter (DOM) derived from organic amendments on HMs and microbial community dynamics in waste slag are still unclear. Here, the effect of DOM derived from organic amendments (cow manure) on the geochemical behaviour of HMs and the bacterial community dynamics in mercury (Hg)-thallium (Tl) mining waste slag were investigated. The results showed that the Hg-Tl mining waste slag without the addition of DOM continuously decreased the pH and increased the EC, Eh, SO 4 2− , Hg, and Tl levels in the leachate with increasing incubation time. The addition of DOM significantly increased the pH, EC, SO 4 2− , and arsenic (As) levels but decreased the Eh, Hg, and Tl levels. The addition of DOM significantly increased the diversity and richness of the bacterial community. The dominant bacterial phyla (Proteobacteria, Firmicutes, Acidobacteriota, Actinobacteriota, and Bacteroidota) and genera (Bacillus, Acinetobacter, Delftia, Sphingomonas, and Enterobacter) were changed in association with increases in DOM content and incubation time. The DOM components in the leachate were humic-like substances (C1 and C2), and the DOC content and maximum fluorescence intensity (F Max ) values of C1 and C2 in the leachate decreased and first increased and then decreased with increasing incubation time. The correlations between HMs and DOM and the bacterial community showed that the geochemical behaviours of HMs in Hg-Tl mining waste slag were directly influenced by DOM-mediated properties and indirectly influenced by DOM regulation of bacterial community changes. Overall, these results indicated that DOM properties associated with bacterial community changes increased As mobilization but decreased Hg and Tl mobilization from Hg-Tl mining waste slag.
Keywords Hg-Tl mining waste slag · Heavy metal(loid)s · Dissolved organic matter · Bacterial community

Introduction
There are rich nonferrous metal mineral resources in Guizhou Province, China. Large-scale nonferrous metal mining and smelting activities in this region have a relatively long history (Luo et al., 2019a;Xiao et al., 2004), which has left behind a large volume of mine tailing and metal smelting waste slag and has had caused a serious impact on the environmental quality around the mining area (Li, 2006;Meng et al., 2021;Zhang et al., 2020;Zhou et al., 2020b). A typical mercury (Hg)-thallium (Tl) mining area in Lanmuchang, located in southwestern Guizhou, China, has been widely studied in recent years due to highly toxic HMs, such as Tl, Hg, and arsenic (As), contaminating soil and water in this area (Lin et al., 2020;Qiu et al., 2006;Xiao et al., 2004). The continues release of HMs from the waste slag generated from mining activities, without any safety measures, is the main source of HMs in the soil and water environment around the mining area.
Phytostabilization is an important way to control the release and migration of HMs from mine tailings in situ (Luo et al., 2018b;Zhu et al., 2022). Generally, mine tailings present particularly unfavourable conditions, such as poor physical structure and water holding capacity, nutrient and organic matter deficiency, combined with elevated levels of HMs, for vegetation growth (Wang et al., 2017a;Zhu et al., 2022). The addition of organic amendments is a critical step for the success of phytostabilization of mine tailings, which is beneficial for improving the habitat conditions in mine tailings and facilitating the establishment of vegetation (Mendez & Maier, 2008;Wang et al., 2017a;Zhu et al., 2022). Organic amendments such as livestock manure and its compost, sludge, and biogas are often used to improve the physical structure, increase nutrient levels, enhance microbial activity, and reduce the bioavailability of HMs in mine tailings (España et al., 2019;Luo et al., 2022;Risueño et al., 2021;Tian et al., 2020;Wu et al., 2019). The application of organic amendments facilitates an increase in DOM content (Eckley et al., 2021;Robertson et al., 2020;Wu et al., 2021;Xue et al., 2022), as organic amendments are easily decomposed by heterotrophic and organophilic microbes. Organic amendments significantly affect the composition and chemodiversity of DOM and its associations with soil microbial communities Wu et al., 2021). Several studies have indicated that DOM significantly affects the geochemical behaviours of HMs (Geng et al., 2020;Guo et al., 2020;Liu et al., 2020;Wang et al., 2021) through the effects of complexation, adsorption, and changes in redox potential, mineral composition, and microbial community Kida et al., 2019;Liu et al., 2019b;Luo et al., 2018a). Numerous studies have indicated that DOM enhances the mobilization of As (Deonarine et al., 2021;Qian et al., 2022), dissolved THg (Eckley et al., 2021;Li et al., 2019), and Tl (Rinklebe et al., 2020) in abandoned mine sites and contaminated soil. These studies mainly examined the effect of DOM alone on HM mobilization. Previous studies have indicated that the presence of interaction effects between the microbial community and the contents and components of DOM (She et al., 2022;Wang et al., 2021;Wu et al., 2018). The application of organic amendments significantly changes the microbial community in mine tailings (Asemaninejad et al., 2020;Risueño et al., 2021;Wei et al., 2019;Zhou et al., 2020a) because the DOM in organic amendments provides soluble organic substrates that support and sustain heterotrophic microbial communities. Microbial community changes are an important factor for the transformation of HMs (Guo et al., 2021;Jiang et al., 2020;Liu et al., 2019a;Wang et al., 2021). Additionally, the DOM associated with the microbial community significantly influences the mobilization of HMs in soil (Liu et al., 2019c;Qu et al., 2019;Wang et al., 2021;Yan et al., 2020b). Several studies have suggested that there are great differences in the physicochemical and microbial properties between soil and mine tailings (Luo et al., 2018b;Wu et al., 2019Wu et al., , 2021Zhu et al., 2022). However, the effect of the DOM associated with the microbial community changes on the mobilization of HMs in mine tailings remains poorly understood.
To fill the above-mentioned knowledge gap, Hg-Tl mining waste slag containing several DOMsensitive HMs (e.g., Tl, Hg, and As) was selected as the research object. We conducted incubation experiments with the addition of different amounts of DOM derived from commonly used organic amendments (cow manure) to Hg-Tl mining waste slag and measured the variation of contents and fractions of HMs, the contents and components of DOM, and the bacterial community composition with increasing incubation time. The objectives of this study were 1) to investigate the effect of DOM on the dynamics of HMs and the bacterial community in Hg-Tl mining waste slag, 2) to identify the variation characteristics of the content and components of DOM, and 3) to reveal the interaction between DOM-bacterial community-HMs. It is hypothesized that the DOM derived from cow manure may (im)mobilize the sensitive HMs such as Hg, As, and Tl in Hg-Tl mining waste slag by affecting the associated governing factors (pH, Eh, DOM properties, and bacterial community). Revealing the effect of DOM derived from organic amendments on the geochemical behaviours of HMs in Hg-Tl mining waste slag would provide a theoretical basis for evaluating the environmental risks of mine tailings amended with organic amendments.

Waste slag sample and DOM preparation
The study area was located in a typical Hg-Tl mining area in Lanmuchang, southwestern Guizhou Province, China (105º30′32″E, 25º31′32″N) (Fig. 1). This mining area has a long history of mining for Hg (~ 350 years) and Tl (since 1990) (Xiao et al., 2004). This area has a subtropical humid climate with an average temperature of 15.2 °C and annual average rainfall of 1315.3 mm. The waste slag samples were collected randomly from the surface (20 cm) of a Hg-Tl mining waste slag site (disposal age ≥ 30 years) in this region. Then, the waste slag subsamples collected from each sample point were evenly mixed into the composite waste slag samples and transported to the laboratory. The waste slag samples were naturally air-dried at room temperature, thoroughly mixed, sieved through a 2 mm sieve and stored for future use. The mineral compositions of the waste slag mainly include quartz, kaolinite, wollastonite, potassium feldspar, and magnetite. The contents of Tl, Hg, As, and Sb in the waste slag were 209. 63, 46.22, 85.75, and 9.01 mg/kg, respectively (Wen et al., 2021b). The sulphur content in the waste slag was approximately 0.6%.
The organic amendment (cow manure) was gathered from a local cow farm located in Guiyang City, Guizhou Province, China. The DOM derived from the cow manure was extracted according to the methods reported by Xiao et al. (2019). Briefly, the cow manure samples were extracted with deionized water at a solid-liquid ratio of 1:10 (w/v, dry weight basis) and shaken in a horizontal shaker at 200 rpm for 24 h at 25 °C. The mixture was centrifuged at 11,000 rpm for 10 min, and the supernatant was filtered through 0.45-μm membrane filters to obtain the DOM solution, which was stored in a refrigerator at 4 °C until use. Prior to the incubation experiment, the physicochemical properties of the DOM solution were measured. The content of dissolved organic carbon (DOC) in the DOM solution was 639.36 mg·L −1 , and its components were humic-like substances (C1 and C2). Additionally, the pH, redox potential (Eh), and electrical conductivity (EC) were 8.29, −76 mV, 1411 μS·cm −1 , respectively; however, the Hg, Tl, and As contents were not measured in the extracted DOM solution.

Incubation experiment
Forty grams of waste slag sample was weighed and placed in a 500-mL polyethylene plastic bottle, and then 20, 40, 60, and 80 mL of DOM solution extracted from cow manure and the appropriate amount of deionized water were added to maintain a solid-liquid ratio of 1:10 (w/v). The concentration gradients of DOM were 5%, 10%, 15%, and 20%, respectively. Additionally, the waste slag treated with deionized water rather than DOM was used as a control treatment (CK). Thus, a total of 5 treatments were designed, with three parallels per treatment. All the above treatments were conducted by batch incubation for 40 days under constant temperature conditions (25 ± 1 °C). Notably, these incubation experiments were conducted under aerobic conditions. The waste slag residue and its leachate were destructively collected at 0, 10, 20, 30, and 40 days, respectively. The leachate was filtered through a 0.45-μm filter membrane to measure its physicochemical properties. The waste slag samples were divided into two parts: one part was used to determine the HM fractions and functional groups, and the other part was used to determine the composition and diversity of the bacterial community.

Analysis of the physicochemical properties in the leachate
The pH, EC, and Eh of leachate were determined by the electrode method. The content of sulfate (SO 4 2− ) in the leachate was measured using barium chromate spectrophotometry. The As and Hg contents in the leachate were measured using an atomic fluorescence spectrometer (AFS-8510, Haiguang, Beijing, China), and the Tl content in the leachate was measured using an atomic absorption spectrometer (TAS-990, Persee, Beijing, China). The DOC content in the leachate was measured using a total organic carbon (TOC) analyser (TOC-2000, Yuanxi, Shanghai, China).
Three-dimensional excitation-emission matrix spectroscopy (3D-EEM) of DOM in the leachate was performed using a fluorescence spectrometer (F-380A, Gangdong, Tianjin, China). The fluorescence intensity at excitation wavelengths (Ex) of 200-500 nm and emission wavelengths (Em) of 200-600 nm was measured for excitation emission matrix spectroscopy (EEMs). The scanning internals of the excitation wavelength and emission wavelength were 5 nm. Fluorescence scanning used ultrapure water (Milli-Q) to subtract the blanks. Parallel factor (PARAFAC) analysis was conducted by using the DOMFluor toolbox in MATLAB software according to previous studies (Murphy et al., 2013;Stedmon & Bro, 2008). Residual analysis and split-half analysis were used to verify the best allocated components after determining discrete values. The maximum fluorescence intensity (F Max ) (R.U.) was used to estimate the relative levels of individual components (Xu et al., 2021).

Analysis of the functional groups and HM fractions in waste slag
The functional groups of waste slag were measured using a Fourier transform infrared (FTIR) spectrometer (FTIR-850, Gangdong, Tianjin, China). Briefly, an appropriate dried waste slag sample powder was mixed thoroughly with KBr (1:150 sample: KBr ratio) in an agate mortar and finely ground. The mixture was then pressed into a pellet to determine the signal in the range of 400-4000 cm −1 using FTIR. A modified BCR-sequential extraction procedure was used to extract the acid-soluble (F1), reducible (F2), oxidizable (F3), and residual (F4) fractions of HMs (As, Hg, and Tl) in waste slag. In brief, 0.11 mol·L −1 HAc and 0.5 mol·L −1 NH 2 OH·HCl were used to extract the acid-soluble and reducible fractions of HMs, respectively; H 2 O 2 and 1 mol·L −1 NH 4 OAc were added to extract the oxidizable fractions of HMs; and finally, the waste slag residue was digested to obtain the residual fractions of HMs. All samples were processed with reagent blanks and parallel samples at the same time.

Analysis of bacterial community composition and diversity in waste slag
A total of 18 fresh waste slag samples from the CK and DOM (5% and 10%) treatments at 0, 20, and 40 days were selected to determine the composition and diversity of the bacterial community. Considering the heterogeneity of waste slag, three replicate waste slag samples from different treatments were completely mixed into two replicate waste slag samples. The genomic DNA of each waste slag sample was extracted using the E.Z.N.A.® Soil DNA Kit (Omega Bio-Tek Inc., Norcross, GA, USA) according to the manufacturer's protocols. DNA concentration and purity were checked with TBS-380 (TurnerBio-Systems, USA) and NanoDrop 2000 (Thermo Scientific, USA), respectively. The quality of extracted DNA was examined by 1% agarose gel electrophoresis. The hypervariable V3-V4 regions of 16S rRNA genes were subsequently amplified by polymerase chain reaction (PCR) using the primer pair 338F (5'-ACT CCT ACG GGA GGC AGC AG-3') and 806R (5'-GGA CTA CHVGGG TWT CTAAT-3') with barcodes. The PCR amplification of 16S rRNA genes was performed according to previous study (Luo et al., 2018b). Sequencing was performed based on an Illumina MiSeq platform (San Diego, CA, USA) by Majorbio BioPharm Technology Co. Ltd. (Shanghai, China). The raw sequencing data were uploaded to the NCBI Sequence Read Archive (SRA) database (Accession Number: PRJNA823598). Sequencing data analysis processes according to the method described by previous study (Luo et al., 2018b).
The indices of Chao1, ACE, Shannon-Wiener, and Simpson were calculated to evaluate the richness and diversity of the bacterial community based on OTUs using the MOTHUR program.

Statistical analysis
All statistical analyses were performed using SPSS 25.0. The results are expressed as the means ± SD (standard deviation). Differences in the diversity indices of the bacterial community in waste slag and the physicochemical properties of leachate in the different treatments were analysed using one-way ANOVA and Duncan's multiple comparison tests, and p < 0.05 was considered significant. Graphs were drawn using the R package version 3.4.1 and Origin 2020.

Changes in physicochemical properties in the leachate
The presence of DOM significantly influenced the physicochemical properties of the leachate (Fig. 2). The Eh and EC values and SO 4 2− content in the CK treatment increased with increasing incubation time; however, the pH showed the opposite trend. This result indicated that the Hg-Tl mining waste slag was gradually oxidized when the waste slag came in contact with water and oxygen, which led to the properties of leachate showing low pH and high EC, Eh, and SO 4 2− content. These results are consistent with those of previous studies (Wen et al., 2021a(Wen et al., , 2021b. The oxidization of sulfide minerals in Hg-Tl mining waste slag to continue releasing H + , SO 4 2− , and HM ions is the main reason for the increase in the EC of leachate (Wen et al., 2021a(Wen et al., , 2021b. Compared to the CK treatment, the addition of DOM significantly facilitated an increase in pH and EC values and SO 4 2− content and a reduction in Eh value in the leachate. The degree of the effect of DOM on the pH, EC, and Eh values and SO 4 2− content in the leachate significantly increased with increasing DOM content. Additionally, the Eh and EC values and SO 4 2− content in the same DOM content treatment increased with increasing incubation time, but the pH showed the opposite trend. The increased pH of leachate in the presence of DOM may be due to the DOM buffering effect on leachate pH (Luo et al., 2018a). The Eh of leachate in the presence of DOM decreased significantly, and the Eh was the lowest when DOM was added in the maximum amount, which indicated that DOM rich in organic matter decomposition could consume oxygen in the leachate (Beauchemin et al., 2018;Luo et al., 2018a;Rakotonimaro et al., 2021). Therefore, the oxidation of sulfide minerals in the Hg-Tl mining waste slag was reduced to a certain extent. However, the content of SO 4 2− continued to increase with increasing incubation time, which indicated that DOM decomposition could not consume all of the oxygen entering the leachate, and the leachate still remained an oxidative state, resulting in the continued oxidation of sulfide minerals in Hg-Tl mining waste slag. The increase in EC values of leachate in the DOM treatments might be attributed to the DOM itself contained abundant ions and the DOM promoted the weathering of minerals in the waste slag caused the release of HM ions. Luo et al. (2018a) suggested that an organic solution long-term cover on high-sulphur coal gangue significantly decreased the Eh values and the SO 4 2− content. Whether long-term coverage of high concentrations of DOM on Hg-Tl mining waste slag can significantly reduce Eh and SO 4 2− content remains to be further studied.

Changes in the content and composition of DOM in leachate
The variation in the content and composition of DOM in the leachate is shown in Fig. 3. The content and Fig. 2 The effect of DOM on the physicochemical properties in the leachate of Hg-Tl mining waste slag. Columns with the same lowercase letter(s) indicate no significant differences among the treatments of different concentrations of the same DOM at p < 0.05, and the same capital letter(s) indicate no significant differences among the treatments of the same DOM concentration at different incubation times at p < 0.05 Fig. 3 The variation in the contents and components of DOM in the leachate. a, b, c, and d represent the contour plots and maximum fluorescence value of excitation and emission spectra for the two modelled components, respectively; e represents the DOC content in the leachate of different treatments. Columns with the same lowercase letter(s) indicate no significant differences among the treatments of different concentrations of the same DOM at p < 0.05, and the same capital letter(s) indicate no significant differences among the treatments of the same DOM concentration at different incubation times at p < 0.05 composition of DOM were not determined in the CK treatment. The higher the DOM concentration was, the higher the DOC content. The DOC content of the leachates in the different DOM treatments gradually decreased with increasing incubation time. Decreases in DOC content may result in related reactions, including mineral-associated adsorption/coprecipitation and microbe-induced degradation (Coward et al., 2018;Lee & Hur, 2020;Li et al., 2021b;Traving et al., 2017). Wen et al. (2021a) suggested the presence of the primary minerals (e.g., quartz and anatase) and the secondary minerals (e.g., kaolinite and jarosite) in Hg-Tl mining waste slag. This result indicated that these primary and secondary minerals could adsorb DOM and prevent it from being transported upwards to the overlying water. The DOM in the leachate contains component 1 (C1) (Ex/Em 340/410 nm) and component 2 (C2) (Ex/ Em 400/450 nm). Several studies have suggested that these components are humic-like substances (Chen et al., 2003;Xu et al., 2021). This indicated that the DOM components in the leachate were dominated by humic-like substances. Although C1 and C2 represent similar substances, their fluorescence peak positions are different. This result may be due to the changes in the concentration of DOM during the incubation period, which are caused by red or blueshifts between the components. Generally, humic-like components (C1 and C2) are related to the stability of DOM. The formation and degradation process of humic-like substances can be divided into biological and/or nonbiological processes, and microbial regulation is a particularly important process (Qi et al., 2019). Previous studies indicated that the structural changes in DOM are closely correlated with the stability of DOM (Antony et al., 2017;Shao et al., 2009). Microbial degradation of proteins, lignocellulose, and carbohydrates is the main source of humic-like substances (Shan et al., 2019;Wu et al., 2017;Xu et al., 2021). DOM can not only provide an energy source for microorganisms (Shan et al., 2019) but also complex with HMs (Li et al., 2011) because DOM contains abundant organo-functional groups.
The dynamic changes in F Max values of different fluorescent components are shown in Fig. 3. The F Max value of C1 was greater than that of C2 in each treatment. The fluorescence intensities of C1 and C2 ranged from 0-0.20 and 0-0.12 R.U., respectively. The F Max values of C1 and C2 in the higher DOM content treatments were greater than those in the lower DOM content treatments. This result indicated that the fluorescence intensities are mainly related to the concentration of DOM (Wen et al., 2021b). The F Max values of C1 and C2 generally first increased and then decreased with increasing incubation time. The highest F Max values of C1 and C2 occurred at approximately 30 days. This result indicated that the decomposition of C1 and C2 by the microbial community occurred from 30-40 days. A previous study suggested that incubation time is the driving factor for the composition and diversity of DOM molecules (Kellerman et al., 2018). It should be noted that the F Max value of C2 in the 5% DOM treatment gradually decreased until it disappeared with increasing incubation time, which is consistent with the variation in DOC content in the 5% DOM treatment. The results indicated that the C2 in the 5% DOM was more easily decomposed by the microbial community.
HM dynamics in the leachate and their fractions in the waste slag HM dynamics in the leachate HM dynamics in the leachate are shown in Fig. 4. Compared to the CK treatment, the addition of DOM inhibited the release of Hg and Tl but promoted the release of As into the leachate. This result indicated that the chemical properties of different types of HMs have different leaching behaviours. The Hg content in the leachate of the CK and 5% DOM treatments showed a similar trend of first increasing and then decreasing; however, the Hg content in the other treatments increased slowly with increasing incubation time. The Hg content in the leachate of the DOM treatments was lower than that of the CK treatments; additionally, the higher the DOM concentration was, the lower the Hg content in the leachate. The Tl content in the leachate of the CK and 5% DOM treatments significantly increased at 0-10 days and increase slowly in the later period of 10-40 days. The Tl content in the leachate of the DOM treatments was lower than that of the CK treatments; additionally, the higher the DOM concentration was, the lower the Tl content in the leachate. Wen et al. (2021b) suggested that Hg and Tl were dissolved from Hg-Tl mining waste slag without the addition of DOM and released into the aqueous phase under low Fig. 4 The variation characteristics of HM content in leachate and HM fractions in waste slag. a, b, and c represent the variation in the Hg, Tl, and As contents in the leachate of different treatments, respectively; d, e, and f represent the variation in the Hg, Tl, and As fractions in the waste slag of different treatments, respectively; F1 = acid-soluble fraction, F2 = reducible fraction, F3 = oxidizable fraction, F4 = residual fraction pH and high Eh conditions. The lower contents of Hg and Tl in the leachate in the DOM treatments indicated that DOM plays an important role in preventing the migration of Hg and Tl. Antić-Mladenović et al. (2017) demonstrated that high Eh levels enhance Tl mobilization; however, the present study showed that the application of DOM significantly decreased the Eh in the leachates (Fig. 2) and that Tl was significantly negatively correlated with Eh (Fig. 8), which indicated that the application of DOM induced a decrease in the Eh in the leachates and was an important factor that prevented the mobilization of Tl from Hg-Tl mining waste slag. Additionally, several studies have suggested that Tl may exist in the form of colloidal particles and that its binding to DOM to form ligands reduces the content of free cations (Belzile & Chen, 2017;Nagel et al., 2019). Therefore, the colloidal particles of Hg and Tl binding to DOM to form ligands adsorbing on the surface of the waste slag may be an important reason for the decrease in Hg and Tl contents in the leachate.
The As content in the leachate of the CK treatment was not detected. Xiao et al. (2004) suggested that the minerals in Hg-Tl mining waste slag are mainly TlAsS 2 , TlHgAsS 3 , TlFeS 2 , etc. These minerals in Hg-Tl mining waste slag are easily subjected to natural weathering and leaching and result in As release (Wen et al., 2021b), but the As released may be attenuated by secondary weathering products of the waste slag (e.g., iron oxyhydroxides) (Elghali et al., 2021;Murciego et al., 2011). Compared to the CK treatment, the addition of DOM significantly increased the As released from Hg-Tl mining waste slag, and the higher the DOM concentration was, the more As was released. Unlike Hg and Tl, the addition of DOM enhanced As mobilization, which was consistent with a previous study (Deonarine et al., 2021). Previous studies suggested that As and Sb show a similar geochemical behaviour, and As has a strong affinity for humic-like substances (Fan et al., 2019). Deonarine et al. (2021) found that both redox conditions and DOM affect As mobilization, but redox seems to drive As mobilization more than DOM. Mladenov et al. (2015) indicated that DOM has the capacity to promote mobilization of As from iron mineral sorbents by reductive dissolution and ligand exchange. The above results indicated that the addition of DOM derived from organic amendments could increase the As released from Hg-Tl mining waste slag. However, Hammond et al. (2020) indicated that organic amendments associated with phytostabilization practices did not significantly increase As mobilization in pyritic mine tailings. Whether the organic amendmentaided phytostabilization of Hg-Tl mining waste slag sites increases As mobilization in waste slag or not requires to be further investigation.

HM fractions in the waste slag
The HMs fractions of Hg-Tl mining waste slag in the different treatments are shown in Fig. 4. The proportions of Hg, Tl, and As in waste slag in the different treatments were the highest, accounting for 45.54%, 88.26%, and 80.32%, respectively. The proportions of HM fractions followed the order F4 > F3 > F1 > F2 for Hg fractions, F4 > F3 > F1 > F2 for Tl fractions, and F4 > F3 > F2 > F1 for As fractions. It is also worth noting that Hg was mainly in the oxidizable fraction at 0-10 days. The proportion of the residual fraction of HMs was the highest at 20-40 days due to the decrease in the percentage of the acid-soluble fraction of HMs. This is due to the permeability of oxygen and leaching liquids during the leaching process and leads to the dissolution of HMs in the acid-soluble fraction, which provides conditions for the release and migration of HMs from minerals of waste slag (Wen et al., 2021a). It can be observed that the sum of the percentages of the first three fractions of Tl is less than that of the residual fraction, which indicated a lower extractability of Tl in Hg-Tl mining waste slag. The lowest proportions of the acid-soluble fraction of As were observed in the CK treatment, which is an important reason for the low As content in the leachate of the CK treatment. Additionally, the As in waste slag occurred in the residual fraction, which indicated that the As in the waste slag was more stable.

FTIR spectrum of waste slag in the presence of DOM
The FTIR spectrum of waste slag is shown in Fig. 5. Two distinctive doublet bands were found at approximately 3390,1624,1086,1028,798,778,690,514, and 467 cm −1 . The doublet bands at approximately 797/778 cm −1 and 1086/1028 cm −1 are assigned to symmetric Si-O and asymmetric Si-O stretching, respectively (Luo et al., 2019b;Wen et al., 2021a;Yin et al., 2018;Zhang et al., 2022). The bands at 1 3 Vol.: (0123456789)  (Frost et al., 1998;Luo et al., 2019b;Zhang et al., 2022). The peaks at 3695/3650 and 3617 cm −1 correspond to the stretching vibrations of surface −OH and internal −OH, respectively (Madejová et al., 2002). These absorption peaks are similar to those of quartz and kaolinite (Zhang et al., 2022), which is due to the mineral composition of Hg-Tl mining waste slag mainly including quartz and kaolinite (Wen et al., 2021a). In addition, the special absorption peaks of 626/1160 and 1871 cm −1 in the Hg-Tl mining waste slag are assigned to the SO 4 2− bending vibrations and the −C = O stretching vibrations, respectively (Wen et al., 2021a;Yin et al., 2018). The characteristic absorption peaks in the different treatments were slightly shifted, but no absorption peaks were generated or disappeared. This result indicated that the addition of DOM has little impact on the functional groups of waste slag, which may be due to the complex composition of waste slag, its primary minerals being relatively stable, and different concentrations of DOM having no obvious effect on the functional groups of waste slag in a short time. However, some differences were observed for the intensity of the major bands between CK and DOM treatments at different inoculation times. The intensity of the characteristic absorption peak at 1086/1028 cm −1 in the CK treatment decreased with the increase of inoculation time, which might be attributed to the fact that the acidic conditions are conducive to the dissolution of silicate minerals.

Bacterial community diversity of waste slag
The richness, evenness, and diversity indices of the bacterial community in waste slag are shown in Table 1. Compared with the CK treatment, the ACE and Chao1 indices in the DOM treatments were significantly increased, and the Simpson and Shannon indices were significantly decreased and increased, respectively. The results showed that the richness and diversity of the bacterial community in the waste slag increased after the addition of DOM. There were no significant differences in the Simpson and Shannon indices between the DOM treatments, but there were significant differences in the ACE and Chao1 indices at 0 days. The higher the DOM content was, the higher the richness and diversity of the bacterial community. This result indicated that the addition of DOM could be beneficial for increasing diversity and richness of the bacterial community because DOM can provide carbon sources for bacterial communities. Additionally, the richness and diversity of the bacterial community in all waste slag samples decreased Table 1 Variation characteristics of the alpha diversity of the bacterial community in different treatments † CK, 5%, and 10% represent the waste slag treated with deionized water and 5% and 10% DOM, respectively. ‡ Columns with the same lowercase letter(s) indicate no significant differences among the treatments of different concentrations of the same DOM at p < 0.05, and those with the same capital letter(s) indicate no significant differences among the treatments of the same DOM concentration at different incubation times at p < 0.05.

Sample
Diversity , released from waste slag to the leachate, adverse to the abundance and diversity of the bacterial community Zhao et al., 2019). Good's coverage estimator > 99% suggested that most of the bacterial OTUs in each waste slag sample were well captured.

Bacterial community diversity of waste slag
The change in the bacterial community composition of waste slag in the different treatments is shown in Fig. 6. The dominant bacterial phyla were Proteobacteria, Firmicutes, Acidobacteriota, Actinobacteriota, and Bacteroidota in the different treatments (Fig. 6a). This result indicated that these bacterial phyla were tolerated in the Hg-Tl mining waste slag. There were significant differences in the relative abundances of Proteobacteria, Firmicutes, Acidobacteriota, Actinobacteriota, and Bacteroidetes in the different treatments (Fig. 6b). Compared with the CK treatment, the addition of DOM greatly changed the relative abundance of Firmicutes, Acidobacteriota, Actinobacteriota, and Bacteroidetes. The relative abundance of Proteobacteria was more than 70% at 20 days, which indicated that the bacterial phylum became the dominant metal-tolerant bacterial phylum Yan et al., 2020a;Zhao et al., 2019). The relative abundance of Bacteroidota reached the highest value (19%) in the 10% DOM treatment at 20 days. Wang et al. (2018) found that Proteobacteria and Bacteroidetes play a major role in organic matter Fig. 6 Variation in the relative abundance of the bacterial community at the phylum level. a Represents a circos plot of the relationship between bacterial species at the phylum level and different treatments. The left semicircle represents species composition at the phylum level, the outer circle colours represent different reaction times, the middle circle colours represent groups of species, and the inner circle colours represent the relative abundance of different species. The right semicircle represents the distribution of different taxa at the phylum level, the colour of the outer circle represents different species, and the colour of inner circle represents the proportion of different species at the phylum level. b Represents significant differences in the relative abundances of the top 5 bacterial phyla in the waste slag of different treatments. The length of the colour blocks represents the average relative abundance of different species, and the colour columns with different colours represent different treatments. * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001 degradation and C cycling. The decrease in the relative abundance of Firmicutes and Actinobacteria was attributed to oxidation of the minerals in the mine tailings (Chen et al., 2014). The relative abundance of Firmicutes and Actinobacteria increased at 0 days, possibly due to the addition of DOM. With increasing incubation time, however, the relative abundance of Firmicutes and Actinobacteria decreased, which may be attributed to competition from other bacteria (e.g., Acidobacteria, Proteobacteria, and Bacteroidetes). The relative abundance of Proteobacteria was the largest, but the relative abundance of Firmicutes, Acidobacteriota, and Actinobacteriota significantly changed at 40 days. This result indicated that the bacterial community was gradually replaced by dominant species during bacterial community succession until the bacterial community composition reached a relatively stable state. The relative abundance of Acidobacteriota in the CK treatment increased with increasing incubation time, but it was lower than that in the DOM treatments. This result indicated that the relative abundance of Acidobacteriota increased significantly, enhancing the oxidation of Hg-Tl mining waste slag, and the addition of DOM promoted the oxidation effect.
The dominant genera were Bacillus, Acinetobacter, Delftia, Sphingomonas, and Enterobacter in the waste slag (Fig. 7a, b). The relative abundances of Acinetobacter, Delftia, Sphingomonas, and Enterobacter in the CK treatment were higher than those in the DOM treatments at 0 days; however, the relative abundance of Bacillus showed the opposite Fig.7 Variation in the relative abundance of the bacterial community at the genus level. a Represents the circos plot of the relationship between bacterial species at the genus level and different treatments. The left semicircle represents species composition at the genus level, the outer circle colours represent different reaction times, the middle circle colours represent groups of species, and the inner circle colours represent the relative abundance of different species. The right semicircle represents the distribution of different taxa at the genus level, the colours of the outer circle represents different species, and the colour of the inner circle represents the proportion of different species at the genus level. b Represents significant differences in the relative abundances of the top 5 bacterial genera in the waste slag of different treatments. The length of the colour columns represents the average relative abundance in different species, and the colour columns with different colour represent different treatments. * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001 variation trend. This result indicated that the addition of DOM decreased the relative abundance of Acinetobacter, Delftia, Sphingomonas, and Enterobacter but increased the relative abundance of Bacillus. The relative abundances of Bacillus, Acinetobacter, Delftia, Sphingomonas, and Enterobacter in all treatments decreased with increasing incubation time. A previous study suggested that Bacillus is a rhizosphere growth-promoting bacterium that plays an important role in the phytoremediation of HM-contaminated sites (Narayanan et al., 2021;Nayak et al., 2019). The addition of DOM increased the relative abundance of Bacillus at 0 days, but its abundance decreased with the increase in pollutants released from waste slag. This result indicated that the change in physicochemical properties in the leachate was unfavourable for the Bacillus growth. Several studies have suggested that Acinetobacter, Delftia, Sphingomonas, and Enterobacter play important roles in As(III) oxidation and reduce its mobility (Biswas et al., 2019;Karn & Pan, 2016;Li et al., 2016;Shi et al., 2013). The present study showed that the As content in the leachate of the CK treatment was lowest, but that in the DOM treatments was highest. This is probably because Acinetobacter, Delftia, Sphingomonas, and Enterobacter enhanced the oxidation of As(III) and reduced its mobility in Hg-Tl mining waste slag without DOM addition, but DOM addition suppressed the activity of Acinetobacter, Delftia, Sphingomonas, and Enterobacter, causing an increase in the As release. Previous studies suggested that the most abundant OTUs of Sphingomonas were present in slightly and extremely acidic lead/zinc mine tailings (Chen et al., 2013), which were resistant to As and Tl (Deng et al., 2020;Rasool & Xiao, 2018;Rasool et al., 2020) and played an important role in the weathering of Tl-rich sulfides and biogeochemical cycling of Tl (Rasool & Xiao, 2018;Rasool et al., 2020). This study showed that the decrease in the Tl released may be attributed to the fact that the activity of Sphingomonas was suppressed along with the increase in pH of the leachate after DOM addition.

Correlation between environmental factors and bacterial community
Spearman correlation coefficient analysis was performed by comparing the major environmental factors with the bacterial community (Fig. 8). Proteobacteria was significantly negatively correlated with pH, DOC, and F Max (C2) in the leachate. This result indicated that Proteobacteria has an important contribution to consuming the DOC and changing the DOC components. A previous study also suggested that Proteobacteria play an important role in the degradation of organic matter and promotion  (Wang et al., 2018). Acidibacillus and Acidiphilium were two bacterial genera of Acidobacteriota that were significantly negatively correlated with pH and significantly positively correlated with Eh, SO 4 2− , Tl, and Hg in the leachate. Several studies have suggested that Acidibacillus and Acidiphilium are important bacterial genera for sulfide mineral oxidation (Gupta et al., 2021;Huang et al., 2021;Liu et al., 2019b). Figure 5 also shows that the addition of DOM increased the relative abundance of Acidobacteriota. These results indicated that the increased relative abundance of Acidobacteriota was an important reason for the oxidation of sulfide mineral in Hg-Tl mining waste slag and the release of SO 4 2− , Tl, and Hg. Sulfuriferula was significantly positively correlated with SO 4 2− , Tl, Eh, and EC in the leachate. The Sulfuriferula genus belongs to Betaproteobacteria and is a novel Fe-and S-oxidizing microbe (Liu et al., 2019b). Several studies have suggested that Sulfuriferula can not only autotrophic growth on inorganic sulphur compounds but also heterotrophic growth on various organic substrates (e.g., organic acids, sugars, and complex organic substrates) (Jones et al., 2017;Watanabe et al., 2016). The results suggest that Sulfuriferula in leachate with a high content of Tl can still retain resistance and metabolic activities.
As was significantly positively correlated with Burkholderia-C-P, Occallatibacter, Terriglobus, and Ralstonia. Li et al. (2021c) suggested that Burkholderia plays a potential role in As (III) oxidation. Burkholderia was isolated from highly contaminated soil from As and lead smelters (Sultana et al., 2012), confirming its resistance to As (III). Several studies have indicated that Ralstonia species are As-and Tl-tolerant bacteria (Rasool et al., 2020;Wang et al., 2017b), and they have also been reported to be able to reduce As(V) (Mergeay et al., 2003;Mondal et al., 2008) and respire using As under anoxic conditions (Vaxevanidou et al., 2015). The DOC was significantly positively correlated with Ralstonia, indicating that the environmental nutrient level plays an important role in promoting the growth of the genera. Some studies have indicated that Ralstonia can also remove Tl contamination through the oxidization of Tl(I) free ions to Tl 2 O 3 and/or dimethylthallium (Liu et al., 2019b). Pseudomonas was significantly negatively correlated with Hg and Tl. Mello et al. (2020) indicated that Pseudomonas promotes mercury volatilization. This result indicated that although the presence of Acidibacillus, Acidiphilium, and Sulfuriferula promotes the oxidation of sulfide minerals in Hg-Tl mining waste slag and the release of Hg and Tl, the presence of Pseudomonas and Ralstonia promotes Hg volatilization and oxidization of Tl(I) free ions to Tl 2 O 3 and/or dimethylthallium, leading to a decrease in the contents of Hg and Tl in the leachate. The correlations between HMs and DOM content/ components and the bacterial community showed that the addition of DOM promotes changes in the bacterial community and decreases the release of Hg and Tl but increases the release of As.

Conclusion
The Hg-Tl mining waste slag in the CK treatment continuously decreased the pH and increased the EC, Eh, SO 4 2− , Hg, and Tl values in the leachate with increasing incubation time. The addition of DOM significantly increased the pH, EC, SO 4 2− content and As content but decreased the Eh, Hg content, and Tl content compared to the CK treatment. The diversity and composition of the bacterial community were changed in association with the DOM content and incubation time. The DOC content of the leachate in the DOM treatments decreased with increasing incubation time. The DOM in the leachate included two components (C1 and C2), which were humic-like substances; the maximum fluorescence intensity (F Max ) values of C1 and C2 in the leachate first increased and then decreased with increasing incubation time. The FTIR spectrum showed that the transformation of HMs may not be influenced by the functional groups of DOM adsorbed on the waste slag. The correlations showed that the geochemical behaviours of HMs in waste slag were directly influenced by DOM properties and indirectly influenced by DOM regulation of bacterial community changes.
Author contribution YH designed the study, performed the experiments and analysed the data, and contributed to writingoriginal draft preparation. YL designed the study, analysed the data, and contributed to writing-reviewing and editing. CW performed the experiments. LL performed the experiments. CW performed the experiments. YW contributed to writingreviewing and editing. All the authors read and approved the final manuscript.

Data availability
The data sets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations
Competing interests The authors declare no competing interests.
Ethics approval Not applicable.

Consent to participate Not applicable.
Consent for publication Not applicable.