Introduction

Antimony (Sb) is a toxic and carcinogenic metalloid considered a pollutant of priority interest by the United States Environmental Protection Agency, US-EPA (US-EPA 1979). The permissible limits of Sb in drinking water, as recommended by the USEPA, European Union (EU), and World Health Organization (WHO) are very low, 6, 5, and 20 μg/L, respectively (US-EPA 2009; EU 1998; WHO 2022). Antimony, the ninth-most mined metal worldwide (Scheinost et al. 2006), is finding increasing applications in the electronics industry where it is used to make some semiconductor devices, such as infrared detectors and diodes. Other uses are the production of flame retardants, textiles, PVC stabilizers, adhesives, tires, brake linings, and as a catalyst for the production of polyethylene terephthalate (Ungureanu et al. 2015).

Antimony can be released into the environment through contaminated water produced by mining operations, smelting factories, and other industrial activities (Wang et al. 2013a, b). A reported case of high levels of Sb in water (up to 6.4 mg/L) was found in rivers around the Sb mine at Xikuangshan (China), and the sediments ranged from 57 to 7316 mg Sb/kg (Wang et al. 2011). The concentration of Sb in soils can reach several hundred mg/kg (Gebel et al. 1996). The concentration of Sb in wastewater can vary depending on the location and anthropogenic activities (mining, printing, dyeing, and metal industry), ranging between 2.8 and 60 mg/L (Zhang et al. 2016; Inam et al. 2018; Gan et al. 2023). This metalloid is mostly present in contaminated (waste)water in two oxidation states: antimonite (Sb(III)) and antimonate (Sb(V)). Conventional treatment technologies typically used to remove Sb and its compounds from mine drainage areas include adsorption, coagulation/flocculation, ozone oxidation, membrane separation, solvent extraction, ion exchange, and reductive electrolysis (Wang et al. 2020). These methods consume high amounts of energy and resources and produce residues that need to be treated and disposed, increasing the total cost of the treatment. The biological reduction of Sb(V) to Sb(III) followed by the precipitation of Sb(III) with biogenic sulfide (as Sb2S3) has been reported in Sb mine drainage (Wang et al. 2013a, b). This microbially-mediated biomineralization process can be used as a remediation method for antimony-rich mine drainage (Wang et al. 2020).

Dissimilatory Sb(V) reduction to Sb(III) by prokaryotic microbial communities in anoxic sediments has been observed under anaerobic conditions (Kulp et al. 2014; Abin and Hollibaugh 2014). Abin and Hollibaugh (2014) reported the enrichment and cultivation of microorganisms of the Bacilliales order capable of respiring antimonate as a terminal electron acceptor for anaerobic respiration using lactate as the sole carbon and energy source, and producing microcrystals of antimony trioxide (Sb2O3). Kulp et al. (2014) observed that anoxic sediment microcosms and enrichment cultures also reduced sulfate, and the precipitation of insoluble Sb(III)-sulfide complexes was a significant sink for reduced Sb. These findings indicate the importance of studying complex anaerobic communities to correlate the microbial composition with the reduction and/or precipitation of Sb.

Anaerobic processes are widely used as an economical option to remove biodegradable organic matter and heavy metals from wastewater (Chen et al. 2008; Fu and Wang 2011). However, the fate of Sb in anaerobic wastewater treatment systems is poorly understood. Likewise, little is known about the role of anaerobic microorganisms involved in the reduction of Sb(V) to Sb(III) since the composition of the microbial communities varies from one wastewater treatment plant to another. Then, this is the first study exploring the capability of different microbial consortia to carry out the Sb removal and transformation under anaerobic conditions, which can facilitate the development of anaerobic processes for the bioremediation of Sb-contaminated water (e.g., acid mine drainage and other aqueous streams contaminated by Sb) and the recovery of Sb. This research aimed to study the microbial reduction of Sb(V) by inocula from different sources under anaerobic conditions and correlate the reduction rates observed with the microbial community composition.

For this purpose, six inocula were obtained from different sources (anaerobic pond sediments, return activated sludge, anaerobic sludge obtained from full-scale up-flow anaerobic sludge blanket (UASB) bioreactor systems treating recycled paper wastewater, anaerobic sludge obtained from a full-scale UASB reactor treating brewery wastewater, and anaerobically digested sludge). Experiments were incubated in batch mode with the presence of a defined Sb concentration. To determine the Sb removal and Sb(V) reduction, the amount of total Sb, Sb(V), and Sb(III) was monitored. Then, the composition of the microbial communities was determined using the Illumina MiSeq platform. Finally, a principal components analysis (PCA) was carried out to determine relationships between the microbial community and Sb performance.

Materials and methods

Chemicals

Sb(V) and Sb(III) were supplied as potassium hexahydroxoantimonate (V) (CAS # 12208-13-8, purity ≥ 99.0%) and potassium antimony tartrate (CAS# 331753-56-1, purity ≥ 99.95%), respectively. Both chemicals were obtained from Sigma-Aldrich (St. Louis, MO, USA).

Inocula source

Microbial reduction of Sb(V) was tested using six different inoculum sources, namely, anaerobic pond sediments (AC, Agua Caliente Park, Tucson, USA); return activated sludge (AN, Agua Nueva Water Reclamation Facility, Tucson, AZ, USA), anaerobic sludge obtained from full-scale up-flow anaerobic sludge blanket (UASB) bioreactor systems treating recycled paper wastewater (R, Rocktenn, Syracuse, New York, USA and E, Eerbeek, The Netherlands), anaerobic sludge obtained from a full-scale UASB reactor treating brewery wastewater (M, Mahou, Guadalajara, Spain), and anaerobically digested sludge (INA, Ina wastewater treatment plant, Tucson, AZ, USA). The different inocula were stored at 4 °C.

Reduction batch assays

Batch experiments were conducted in 160-mL glass serum bottles (Wheaton, Millville, NJ, USA) with a working volume of 100 mL, and 60 mL of headspace. The inoculum concentration was 1.5 g volatile suspended solids (VSS) per liter of the previously described sources. Mineral basal medium (in mg/L): K2HPO4 (12.5), CaCl2.2H2O (10), MgCl2.6H2O (mg/L), NH4HCO3 (20), NaHCO3 (2000), yeast extract (10), and 1 mL/L of trace elements solution. The trace element solution contained (in mg/L): H3BO3 (50), FeCl2∙4 H2O (200), ZnCl2 (50), MnCl2∙4H2O (32), (NH4)6Mo7O24∙4H2O (50), AlCl3∙6 H2O (50), CoCl2∙6 H2O (2000), NiCl2∙6 H2O (50), CuCl2∙2 H2O (44), NaSeO3∙5 H2O (100), EDTA (1000), and resazurin (200) was applied based on the proposal by Ramos-Ruiz et al. (2016). The bottles were closed with rubber septa and aluminum crimp seals; the headspace was flushed with N2:CO2 (80:20, v/v) for 3 min to ensure anaerobic conditions, and then they were pre-incubated for 24 h in the dark at 30 ± 2 ºC in an orbital shaker (105 rpm). Subsequently, Sb(V) (19 mg/L) and the electron donor (hydrogen at a stoichiometric excess of 30-fold supplied in the headspace) were added. Duplicated bioassays were incubated at 30 ± 2 °C in an orbital shaker (105 rpm) in the dark. Liquid phase samples were collected periodically for analysis of total Sb, Sb(V), and Sb(III). The samples were centrifuged at 5000 rpm for 15 min and stored at – 4 ºC.

Chemical analysis

The physicochemical parameters as pH and VSS were determined according to the standard methods (APHA, 2005). The total soluble concentration of Sb was measured by inductively coupled plasma-optical emission spectroscopy (ICP-OES, 5100 SVDV Agilent Technologies 5100, Santa Clara, CA, USA) at a wavelength of 217.582 nm. Before the analysis, samples were centrifuged (13,000 rpm, 10 min) and diluted into 2% nitric acid. The removal rate was calculated considering the Sb total obtained from total soluble quantification at the maximal slope elimination of Sb species (µg/L) per day.

Sb speciation (Sb(III) and Sb(V)) was performed using a hyphenated technique coupling liquid chromatography (LC, Waters Millipore 590, Milford, MA, USA) to an ICP-OES instrument, according to Moreno-Andrade et al. (2020). The LC pump was connected to a manual Rheodyne injection valve that was plumbed to two in-line ion chromatography columns (Dionex AG15 4 × 50 mm and Dionex AS15 4 × 250 mm) (Thermo Scientific, Waltham, MA, USA). The effluent from the analytical column was coupled to the liquid inlet to a SeaSpray nebulizer on the Agilent 5100 SVDV ICP-OES using PEEK tubing. The eluent used was ethylenediaminetetraacetic acid (EDTA) 40 mM at a flow rate of 1.0 mL/min. Samples were injected manually in a 300-µL loop. The ICP-OES measurement conditions were: pump speed 40 rpm, read time 1 s, RF power 1.2 kW, nebulizer flow 0.7 L/min, plasma flow 12.0 L/min, auxiliary flow 1.0 L/min. ICP-OES measurements were collected solely in an axial mode for the highest sensitivity. Data for three different Sb emission wavelengths were simultaneously collected (206.834 nm, 217.582 nm, and 231.146 nm) in order to increase confidence in the identities of the peaks. The synchronization between the LC and the ICP-OES runs was done manually. The time-resolved emissions data collected were exported as text (CSV file, integrated using Excel software) according to the methodology proposed by Moreno-Andrade et al. (2020). Calibration curves were obtained with measurements of 12.5, 25, 50, 125, 250, 500, 1000, 2500, and 5000 µg/L for each Sb species. A linear model was selected to fit the dependence of the concentration of the total number of analytes counts. The limits of detection at a wavelength of 217.582 nm for Sb(III) and Sb(V) were 29.5 and 42.4 μg/L. The method´s recovery from samples, including biomass and culture medium, is higher than 96.0% at a wavelength of 217.582 nm for Sb(III) and Sb(V) (Moreno-Andrade et al. 2020). The Sb(V) and Sb(III) final removal, reduction, and rate were calculated considering the data obtained from the speciation methodology.

Microbial community analysis

At the end of each Sb reduction experiment, biomass samples were taken to characterize the microbial communities. Samples were preserved at − 4 °C until analysis. Genomic DNA was extracted from biomass samples using the PowerSoil® DNA isolation kit (MOBIO, Carlsbad, CA, USA) according to the manufacturer’s instructions. The concentration of the DNA was quantified by spectrophotometry using a NANODrop 2000c (Thermo Scientific, Waltham, MA, USA). The DNA was submitted to the Research and Testing Laboratory (RTL, Lubbock, TX, USA) for Illumina MiSeq sequencing, using the bacterial and archaeal universal primers 515F (GTGCCAGCMGCCGCGGTAA) and 806R (GGACTACHVGGGTWTCTAAT) of 16s rDNA gene (Caporaso et al. 2011).

The sequences were analyzed using the QIIME software (Quantitative Insights into Microbial Ecology) (Caporaso et al. 2010). Methodology for dereplication, clustering, selection of OTUs, and taxonomic classification has been described previously (Barragán-Trinidad et al. 2017). Using the identified OTUs, different statistical analysis was run in the R environment (RStudio Team 2016). Principal components analysis (PCA) was carried out to determine relationships between the microbial community at genus level and Sb reduction performance, using the Vegan package (Oksanen et al. 2016). In addition, statistical correlations among members of the identified microbial community and other environmental parameters were determined by Pearson’s coefficient, using the Corrplot package (Taiyun and Simko 2016).

Results and discussion

Microbial reduction of Sb(V)

Table 1 shows the total removal of Sb and Sb(V), and the percent Sb(V) transformed to Sb(III) in the assays with the different inocula. Microbial reduction of Sb(V) to Sb(III) (3–77% reduction) by the inoculum sources tested was observed in anaerobic incubations using hydrogen as an electron donor after a month. The highest reduction rate (12.5 mg Sb/g VSS day) was obtained with the inoculum AN (Fig. 1). Inocula AN and M reduced 84.7 and 100% of the added Sb at 5 and 60 days, respectively. For the other inocula, the Sb reduction percentage ranged from 3 to 50% in 30 days. In addition to Sb reduction, a decrease in the total concentration of Sb in the soluble phase was also observed in the bioassays inoculated with AC, R, E, and INA, suggesting precipitation of Sb(III). Based on these results, the inocula can be divided into two groups: (1) AN and M showing the high rates of Sb(V) reduction, with low removal of total soluble Sb, and (2) AC, R, E, and INA displaying low concentrations of soluble Sb(III) in the medium and high removal of total soluble Sb. The amount of Sb(III) recorded in the liquid phase at the end of the bioassays inoculated with AN, M, and AC ranged from 50 to 102% of the initial Sb (Table 1). In the other inocula, the amount of reduced Sb(III) detected in the liquid medium was lower than the 10% of the added Sb.

Table 1 Removal of total Sb and Sb(V), and Sb(V) reduction under anaerobic conditions by the different inocula tested
Full size table
Fig. 1
figure 1

Kinetics of Sb species using different inocula: A AN, B M, C AC, D R, E E, and F INA

Full size image

The reduction of Sb(V) followed the first-order kinetic reaction equation CSb(V) = Co × exp(−k × t). Then, the Sb(V) reduction rates (k) for AN, M, AC, R, E, and INA were 4.0 × 10–8 (R2 = 0.800), 2 × 10–8 (R2 = 0.920), 8.7 × 10–5 (R2 = 0.837), 6.6 × 10–5 (R2 = 0.781), 3.1 × 10–3 (R2 = 0.915), and 2.7 × 10–6 (R2 = 0.937), respectively.

The Sb transformation observed in all the inocula (Fig. 1 and Table 1) can be associated with the dissimilatory reduction of Sb(V) (Sb(OH)6[−]) to Sb(III) (Sb(OH)30) (Abin and Hollibaugh 2014). In this anaerobic respiration pathway, microorganisms oxidize organic matter using metals as final electron acceptors, unrelated to metal assimilation (Lovley 2002). This process is carried out as an extension of the respiratory chain on the outer surface of gram-negative bacteria, reducing extracellular electron acceptors, for which the electron transport cytochrome proteins are essential. These proteins allow electrons to travel from the cytoplasmic membrane to the periplasm and onto the outer membrane. The outer cytochromes of the membrane have the ability to catalyze the last steps of the respiratory chain by transferring the electrons to the metal (Ritcher et al. 2012). This pathway has been reported and identified for other metalloids such as arsenic. Recent research attributes the microbial reduction of Sb to enzymes belonging to the dimethyl sulfoxide reductase family (Shi et al. 2019a, b; Abin and Hollibaugh 2019). However, this has not yet been confirmed, and there is evidence suggesting that other factors could intervene in the process. For example, it has been observed that the reaction is limited by the accumulation of reduced Sb(III), which could be related to the resistance and tolerance of the microorganisms to Sb(III) (Filella et al. 2007; Li et al. 2016).

The removal of total Sb in AC, R, E, and INA was higher than 61%, while in the bioassays with AN and M, it was 9.3 and 18.4%, respectively (Table 1). Sb removal was considered by the total soluble concentration measured by the ICP-OES method since speciation analysis used in our study has the limitation of possible interferences resulting from the interaction of the analytes with the sample matrix (e.g., sludge, wastewater, etc.), associated to several chemical elements that have a response at wavelengths close to that of Sb which could interfere if present at high concentrations (Moreno-Andrade et al. 2020), resulting in possible variations in the Sb(III) and Sb(V) data. The total concentration of soluble Sb present in the solution at the end of the incubation with some of the inocula (AC, E, R, INA) was considerably lower compared to the initial concentration of Sb(V) added (19 mg/L) (Fig. 2). The decrease in the concentration of soluble Sb is likely due to the precipitation of Sb(III) with biogenic sulfide formed by sulfate reduction (Zhang et al. 2016; Kulp et al. 2014; Wang et al. 2013a, b). Sb(III) forms highly insoluble Sb2S3 salts with sulfide (solubility product = 1.6 × 10–93 M5) (Polack et al. 2009). Sb(III) precipitation with sulfide (e.g., iron sulfide, hydrogen sulfide) is a pH-dependent reaction. A study demonstrated that the reaction of Sb(III) at acidic pH was closely related to the precipitation of sulfide mineral Sb2S3. Meanwhile, basic pH was related to the Sb(III) surface adsorption (Han et al. 2018). In the present study, the initial pH of the assays was 7.0. This explains the simultaneous decrease in the concentration of total soluble Sb and Sb(III) in bioassays inoculated with E, R, INA (Table 1). The drop of the initial Sb in AC, R, E, and INA (Fig. 2) probably occurred by the abiotic reduction of the Sb(V) to Sb(III) with the subsequent precipitation by the presence of residual sulfide in the inocula. It was previously reported that anaerobic sediments and sludges obtained after anaerobic digestion could contain sulfur in the form of organic sulfur, sulfate, and sulfide (Dunnette et al. 1985; Dewil et al. 2008; Yan et al. 2018). The amount of sulfide in the anaerobic sludge can vary based on the type of sludge, and the bioprocess which the sludge was obtained (Yan et al. 2018; Shi et al. 2019a, b).

Fig. 2
figure 2

Removal of total soluble Sb as a function of time in anaerobic bioassays utilizing by different inocula

Full size image

The removal of total antimony, i.e., the decrease of soluble antimony from the aqueous phase, does not only depend on the microorganisms that reduce Sb but also on physicochemical factors such as pH, redox state, and the presence of species that may help to precipitate Sb(III) (e.g., sulfide). Microorganisms can also participate in supplying the species that will later precipitate in the process.

The Sb transformation has been studied in concentrations between 0.5 and 20.0 mg/L of Sb(V) using anaerobic sludge, activated sludge, and paddy soil as inocula (Table 2). The Sb(V) removal ranged from 86.3 to 100.0% (Wang et al. 2013a, b; He et al. 2022). The Sb(V) removal obtained with AN, M, and INA was in the range previously reported. Meanwhile, the Sb(V) removal of AC, R, and E was slightly lower than the reported range. Regarding the Sb(V) reduction to Sb(III), it has been reported between 15.1 and 31.21% (Wang et al. 2013a, b; He et al. 2022). In this study, AC, M, and AN reached 2 to 5.5-fold higher Sb(V) reduction than the reported in previous studies.

Table 2 Comparison of the removal Sb(V) and Sb(V) by inocula from different origin
Full size table

Microbial community analysis

Metagenomic analysis of biomass samples collected at the end of the various bioassays revealed high variability in the microbial communities present in the different inocula tested, probably due to the different origins of the inocula. The inoculum source has been reported to be a critical parameter for the performance of anaerobic processes, which affects the metabolic activity and composition of anaerobic microbial communities (Moreno-Andrade and Buitrón 2004). The species richness among different inocula ranged from 126 to 325, regarding the number of OTUs. Such values are in the low range of bacterial richness previously reported in anaerobic sewage and sludge digestion plants, from 266 to 478 OTU (Buettner and Noll 2018), and in hot springs sediments, from 165 to 1854 OTUs (Wang and Pecoraro 2021). In this sense, the concentration of metals has been suggested to significantly affect the community richness in hot springs (Wang and Pecoraro 2021); and it has been observed that the Sb reduction activity can reduce the microbial diversity from contaminated soil, regardless of the electron donor used (Nguyen et al. 2018).

Figure 3A shows the relative abundance at genus level from the different inoculum. Some inocula showed a clear dominant genus, including the genus Lysinibacillus (44.2%) for Agua Nueva (AN) or the genus Bacillus (50.9%) for Agua Caliente (AC). Other samples showed more than one genus with more than 10% of the OTUs. Mahou (M) showed the presence of the genus Stenotrophomonas (37%), Methanosaeta (12%), and Methanobacterium (10%). For the case of Eerbeek inoculum (E), three genera represented 55% of the total genera present in the sample (Methanobacterium 30.5%, Clostridium 14.6%, and Sphingopyxis 10.2%). Rocktenn (R) showed 21.7% of the sequences related to the Methanosaeta genus with the presence of Pseudomonas (15.8%). INA presented the highest percentage of OTUs unclassified (37.9%), whereas the other samples ranged from 2.4 to 6.7%, and only the genus Pseudomonas reached a value higher than 10%. For the order taxonomic category (Figure S1 supporting material), the three inocula that showed the highest reduction potential were related to the highest percentage of presence of the Bacillaceae family (AC showed 53%, AN 49%, and M 8%). The abundance of order Bacillales increased to 75% in AN community.

Fig. 3
figure 3

A Relative abundance of microorganims at the genus level in the different inocula, and B principal components analysis performed by considering the genus assignment matrix from different inocula. The Sb removal activity in terms of Sb (total—T, and V), and reduction were fit to the spatial distribution. Only the significant correlations are shown (p < 0.05)

Full size image

The multivariate data analysis by PCA of the microbial community composition explained more than 66% of the variation (Fig. 3B). A clear correlation of the community to the Sb removal was observed since total Sb reduction and total Sb removal rate were positively correlated to the community composition (average R2 and p-values of 0.95 and 0.008, respectively), mainly to those samples from anaerobic reactors (INA, E, R, and M). Sb(V) removal and Sb(III) reduction also had a significant correlation with the microbial community structure but with limited collinearity (average R2 value of 0.8).

The Pearson’s matrix showed only a positive correlation of soluble Sb removal with the presence of Simpliscispira and a negative correlation with the Longilinea genus. Inversely, the presence of Longilinea genus was positively correlated with Sb(V) removal rate and reduction to soluble Sb(III). The same positive correlation was observed for Brevibacillus, Sporosarcina, Lysinibacillus genera, belonging to the Bacilli class (Fig. 4). Other studies have reported that the presence of the Bacilli class was positively correlated with the Sb(V) reduction rate but not with the removal of total soluble Sb (Abin and Hollibaugh 2017, 2014). Conversely, a decrease in the presence of the Bacilli class corresponded with low Sb(V) reduction rates, and enhanced removal of total soluble Sb. Further studies about microbial reduction (e.g., a possible correlation between reduction and presence of microorganisms as Bacilli) need to be carried out as a possibility to obtain the Sb reduction or total Sb removal.

Fig. 4
figure 4

Matrix of Pearson’s correlations obtained for the different genus and Sb removal. The Sb (total—T, III, and V), removal (Rem), reduction (Red), removal rate (RemRate), and reduction rate (RemRate) values were considered. Only correlations with p < 0.05 are displayed

Full size image

The composition of the microbial communities in the different inocula can be affected by the presence of Sb, but also by different substrates used as energy and carbon source., e.g., in samples fed with H2, α-Proteobacteria (Rhizobium) was reported (Lai et al. 2016), while the genus Geobacter and Pseudomonas sp. were detected in samples containing sulfate (Zhu et al. 2018). Microorganisms from the genera Alkaliphilus, Clostridiaceae, Tissierella, and Lysinibacillus were also reported in samples obtained near an active Sb mine (Wang et al. 2018). Sinorhizobium sp. isolated near an Sb factory has been reported that can reduce the Sb(V) and produce the biogenic mineral Sb(OH)3 (He et al. 2019; Nguyen and Lee 2014). Simplicispira is a group of denitrifying microorganisms, and this genus has not been previously related to the Sb removal or the Sb(V) reduction. On the other hand, these bacteria have been identified in activated sludges and denitrifying environments where their main contribution is the formation of biofilms due to the exopolysaccharides production (Siddiqi et al. 2020; Chen et al. 2020; Zhang et al. 2022). In this study, Simplicispira might be related to the cell retention of key microorganisms involved in the Sb removal through biofilm formation. Despite such differences, most of the previously cited works evaluated the potential of microbial communities for Sb reduction or removal, using sedimentary samples as inoculum, where different genera from the Proteobacteria phylum have been associated with Sb reduction activity, like Azospira, Dechloromonas, Halodesulfovibrio, Rhizobium, Geobacter, and Desulfobulbus (Lai et al. 2016; Nguyen et al. 2018; Xu et al. 2020; Yang et al. 2021). In this work, microbial communities enriched from engineered environments like M and AN (anaerobic sludge and return activated sludge) presented the higher Sb(V) reduction activity, having the common genera Lysinibacillus as dominant, previously reported in a Sb contaminated soil (Wang et al. 2018). In general, communities obtained from anaerobic environments, like UASB reactors (R, E, M, and INA), had a close ecological distance, possibly explained by the common presence of archaea genera Methanosaeta and Methanobacterium. Also, such anaerobic communities had a higher genera richness compared to AC and AN samples but shared the presence of genera related to Sb removal activity like Pseudomonas, Clostridium, or Lysinibacillus (Wang et al. 2018; Yang et al. 2021; Zhang et al. 2016; Zhu et al. 2018). In this study, the microbial community analysis demonstrated the selection of different bacteria taxons related to Sb removal, even from engineering environments like wastewater treatment plants, either anaerobic (R, E, M, and INA) or aerobic (AN). However, further studies are needed to identify particular microorganisms capable of Sb removal and characterize their metabolic activity.

Conclusions

This study correlates microbial anaerobic consortium from different sources with the antimony (Sb) dissimilatory reduction. Not only Sb reduction was observed, but also a decrease in the total Sb concentration, suggesting the precipitation of Sb(III). It was possible to link the Sb reduction potential with microorganisms from the Bacillaceae family. In some microbial communities, not only the Sb reduction was observed but also a decrease in the total Sb concentration, suggesting the precipitation of Sb(III), which can be helpful for the development of the process for the Sb recovery in anaerobic conditions. In this study, the precipitation of Sb(III) by the presence of sulfide was not evaluated. It has been widely reported the presence of sulfide in anaerobic systems and its contribution to the precipitation of Sb(III). However, this transformation needs to be confirmed in future research. Additionally, the results of the microbial community composition and the Sb removal or transformation could be complemented by their relationship with some operational conditions (e.g., pH, activity of electrons, presence of metals).