Interesting Halophilic Sulphur-Oxidising Bacteria with Bioleaching Potential: Implications for Pollutant Mobilisation from Mine Waste

For many years, research on the microbial-dissolution of metals from ores or waste materials mainly focussed on the study of acidophilic organisms. However, most acidophilic bioleaching microorganisms have limited tolerance to high chloride concentrations, thereby requiring fresh water for bioleaching operations. There is a growing interest in the use of seawater for leaching purposes, especially in regions with less access to fresh water. Consequently, there is a need to find halophilic organisms with bioleaching potentials. This study investigated the bioleaching potentials of four moderately halophilic sulphur-oxidising bacteria: Thiomicrospira cyclica, Thiohalobacter thiocyanaticus, Thioclava electrotropha and Thioclava pacifica. Results revealed T. electrotropha and T. pacifica as the most promising for bioleaching. Pure cultures of the two Thioclava strains liberated about 30% Co, and between 8–17% Cu, Pb, Zn, K, Cd, and Mn from a mine waste rock sample from the Neves Corvo mine, Portugal. Microwave roasting of the waste rock at 400 and 500 °C improved the bioleaching efficiency of T. electrotropha for Pb (13.7 to 45.7%), Ag (5.3 to 36%) and In (0 to 27.4%). Mineralogical analysis of the bioleached residues using SEM/MLA-GXMAP showed no major difference in the mineral compositions before or after bioleaching by the Thioclava spp. Generally, the bioleaching rates of the Thioclava spp. are quite low compared to that of the conventional acidophilic bioleaching bacteria. Nevertheless, their ability to liberate potential pollutants (metal(loid)s) into solution from mine waste raises environmental concerns. This is due to their relevance in the biogeochemistry of mine waste dumps, as similar neutrophile halophilic sulphur-oxidising organisms (e.g., Halothiobacillus spp.) have been isolated from mine wastes. On the other hand, the use of competent halophilic microorganisms could be the future of bioleaching due to their high tolerance to Cl- ions and their potential to catalyse mineral dissolution in seawater media, instead of fresh water.


Introduction
The extraction and processing of mineral resources generate significant amounts of waste globally, mainly tailings and waste rock. Tailings are a slurry of fine-grained residues resulting from the comminution and beneficiation of ores and typically contain secondary precipitates and processing reagents, such as flotation chemicals [1]. Waste rock consists of low-grade bedrock or earth material excavated during mining operations to access the economically valuable ore and is generally coarse, crushed, or blocky, covering a range of sizes, from very large boulders or blocks to fine sand-size particles [2]. These mine wastes are potential threats to the environment, as their exposure to air and water oxidises the metal sulphides and can result in acid generation and metal(loid) mobilisation into the environment. On the other hand, mine wastes are increasingly being considered The mineral composition of NC_01 was determined by a scanning electron microscopebased automated image analysis technique called a Mineral Liberation Analyzer (MLA). The MLA system is able to collect quantitative mineralogical and textural data for samples. The MLA consisted of: (i) a Quanta 650F field emission scanning electron microscope (FE-SEM) (FEI, Hillsboro, OR, USA), (ii) two Quantax X-Flash 5030 energy-dispersive X-ray (EDX) detectors (Bruker, Billerica, MA, USA), and (iii) the MLA software suite version 3.1. To prepare the samples for MLA, 3 g of each sample was mixed with graphite powder and embedded in 25 mm epoxy resin, which was cut vertically, rotated at 90 • C, re-embedded into a second epoxy-resin and then polished to obtain a 30 mm grain mount. The grain mount (containing the sample) was carbon-coated and measured by MLA using the GXMAP measurement mode at horizontal frame width of 500 pixels, 4.78 spot size, probe current of 10 nA, and acceleration voltage of 25 kV.
All chemicals and reagents used in this study were of analytical grade.

Bioleaching Experiments
The bioleaching potentials of the four halophilic sulphur-oxidising bacteria were determined by incubating pure cultures of each organism with the waste rock sample in 250 mL Erlenmeyer flasks, and constantly agitated at 160 rpm at room temperature. The bioleaching experiments were carried out at 5 wt% solid content, using 10% v/v inoculum from cultures containing between 10 9 -10 10 CFU/mL. The same mineral medium used to culture each organism (Section 2.2) was also used for bioleaching. The final volume of the liquid phase in the bioleaching flasks was 100 mL.
For each organism, one-step, two-step and spent medium bioleaching studies were conducted. In one-step bioleaching (1SB), each organism was inoculated into the mineral medium containing the waste rock sample. In two-step bioleaching (2SB), the organisms were allowed to grow to the exponential stage before the addition of the waste rock sample on day 2. In spent medium bioleaching (SMB), the bacterial cells were separated from the liquid media after maximum growth of the organisms prior to using the liquid media as lixiviants for the leaching of the waste rock. The 2SB was conducted to investigate if the contents of the waste rock were toxic to the growth or bioleaching capacity of the organisms. Meanwhile, the SMB was conducted to investigate if the mechanism of bioleaching involved an indirect process that only requires the metabolic products of the organisms.
There was an abiotic control where an equal volume of deionised water substituted the total volume of the mineral medium and the inoculum. Sampling of the leachate was performed every 7 days to monitor the pH (Mettler Toledo), sulphate concentration (Thermo Scientific (San Jose, CA, USA) Dionex Integrion high-performance ion chromatography) and soluble metal(loid) concentrations (Perkin Elmer, ICP-MS). All experiments were run in duplicate for 28 days, and the mean values are reported. Metal(loid) recovery after n days of incubation was calculated as follows: where Mi is the initial mass of the element in the liquid mineral medium, Mn is the mass of the element in the leachate on day-n, and Mt is the total mass of the element in the solid waste sample. At the end of the bioleaching experiments, bioleached residues from promising organisms were harvested, dried at room temperature and then analysed for their chemical and mineralogical content, using the same conditions stated in Section 2.1.

Bioleaching Optimisation Experiments
After the initial screening of the four organisms for bioleaching potential, various processes that may affect the bioleaching activities were studied with the aim of improving the bioleaching activities of the promising organisms. Bioleaching conditions were the same as in Section 2.3, unless otherwise stated. These optimisation processes include: (i) Effect of the Thioclava consortium on the bioleaching of NC_01 T. electrotropha and T. pacifica were grown together to obtain a Thioclava consortium which was used at 10% v/v to conduct 1SB, 2SB and SMB of NC_01 at 5% solid content for 28 days.
(ii) Effect of weekly serial addition of NC_01 and partial nutrient replacement on the bioleaching of NC_01 1SB and 2SB were conducted for the two Thioclava spp. using 10% v/v inoculum. NC_01 was added weekly to obtain 1%, 3% and 5% solid content for week 0, week 1 and week 2, respectively. Half of the leachate was replaced weekly with a fresh mineral medium. This was conducted by centrifuging half of the leachate, discarding the supernatant and resuspending the solids in a fresh mineral medium before transferring it back to the bioleaching conical flasks. The experiment was run for 42 days.
(iii) Effect of high temperature and pressure on the bioleaching of NC_01 The two Thioclava spp. were grown in the ASW medium for 21 days. Then, the spent medium (pH ≈ 3.1) was used as lixiviants for the bioleaching of NC_01 at high temperature (121 • C) and pressure (103,421.4 pa) for 30 min, using an autoclave (Steriltechnik AG, Haldensleben, Germany). After cooling, leachates were analysed for dissolved metal concentrations using the ICPMS.
(iv) Effect of microwave roasting on the bioleaching of NC_01 NC_01 samples were placed in alumina crucibles and heated in a ventilated PYRO advanced microwave furnace (Milestone, Shelton, CT, USA) at 400, 500 and 600 • C, at a ramp time of 30 min and a dwell time of 60 min. The major mineral phases of the microwave-roasted samples were determined using X-ray diffraction (XRD) (PANalytical Empyrean diffractometer, Malvern, United Kingdom [Co-tube, 2 theta 5-80 • with a step size of 0.013 • , 35 kV and 35 mA, PIXcel 3D medipix 1 × 1 area detector]). 1SB and 2SB of the microwave-roasted NC_01 were then conducted with T. electrotropha for 14 days at 5% solid content, using 10% v/v inoculum.

Bioleaching Results of the Four Halophilic Sulphur-Oxidising Bacteria
The concentrations and percentages of metal(loid)s extracted from NC_01 by T. cyclica, T. thiocyanaticus, T. electrotropha and T. pacifica are presented in Tables S1 and A1,  Tables S2 and A2, Tables S3 and A3, and Tables S4 and A4 (Supplementary Material and  Appendix A), respectively. The evolution of pH, total dissolved Fe and sulphate concentrations during the four bioleaching experiments are presented in Figure 1. An increase in sulphate concentration arising from the ability of the four halophilic organisms to oxidise reduced sulphur indicated bacterial growth and activity during the bioleaching experiments. A decreasing solution pH was also used as an indication of sulphur oxidation, and therefore, bacterial growth and activity, as pH reduces with an increasing sulphate concentration [13]. Biomass growth and activity in the presence of the waste rock indicated that the organisms were tolerant to the high metal(loid) contents of the waste rock. The bioleaching capability of the halophilic organisms was assessed based on the quantity of dissolved metal(loid)s (Cu, Pb, Zn, Co, As, Cd, K, Sb, Ag and Mn) in solution resulting from the oxidation of sulphide minerals in NC_01. Generally, for all four organisms, no bioleaching assay (1SB, 2SB and SMB) was observed to be consistently more efficient. The extraction efficiencies for 1SB, 2SB and SMB were comparable for most meta(loid)s.   Sulphur oxidation by T. cyclica produced sulphuric acid, thereby drastically increasing the SO 4 2− concentration and reducing the solution pH within the first 7 days. Specifically, there was a steep increase in sulphate concentration from its initial concentration (~1000 mg/L) on day 0 to 5000 mg/L (for 1SB) and 8535 mg/L (for 2SB) on day 7. The SO 4 2− concentration continued to increase, reaching 9000 mg/L by the end of the 1SB and 2SB experiments. The pH decreased from 10.3 to 9.7 within the first 7 days during the exponential growth phase of T. cyclica and remained relatively constant until the end of the 1SB and 2SB experiments. The oxidation of reduced sulphur compounds during the bioleaching of NC_01 by T. cyclica resulted in metal(loid) dissolution. The total dissolved Fe was also observed to increase within the first 7 days of 1SB and 2SB, from 0.3 to 4.2 mg/L, and continued to increase until around 5.7 mg/L on the last day. By the end of the bioleaching experiment with T. cyclica, (%) 6.0 Cu, 0.6 Pb, 0.8 Zn, 6.2 Co, 44.2 As, 2.3 Cd, 17.4 K, 38.7 Sb, 23.7 Ag, and 0.0 Mn were solubilised from NC_01 via 1SB, 2SB and/or SMB (the highest recovery among the three strategies is stated). However, the maximum recovery for some elements was attained before the last day. For example, Cu and K recoveries reached a maximum (12.5% and 31.2%, respectively) on day-7 via 1SB.
T. thiocyanaticus dissolved (%) 7.3 Cu, 0.4 Pb, 0.1 Zn, 1.3 Co, 0.7 As, 3.3 Cd, 5.2 K, 25 Sb, 14.2 Ag, and 0.3 Mn from NC_01 within the first 7 days. After day 7, the recoveries either remained constant or decreased, except for K, Sb and Ag, which increased to 7.2, 38.7 and 31.8%, respectively, towards the end of the experiment. The pH increased from 7.8 to 9 and 8.5 in 1SB and 2SB, respectively, within the first 7 days, with no major change after this. On the other hand, there was less than a 2-fold increase in sulphate concentrations in the bioleaching assays. The SO 4 2− concentration rose from 2315 to 4560 mg/L in the first 7 days and remained constant until the end of the 2SB experiment. While for 1SB, the SO 4 2− concentration increased gradually throughout the duration of the experiment from 863 mg/L on day 0 to 3915 mg/L on the final day. The pH and SO 4 2− concentration in SMB decreased slightly through the experiment, from 8.6 to 7.6 and 4815 to 3675 mg/L, respectively. The total dissolved Fe decreased considerably within the first 7 days from its initial concentration (0.3 g/L), reaching 0.06 g/L on day 7, and then remained below 0.2 till the end of the 1SB and 2SB experiments.
Bacterial growth and activity led to a marked decrease in pH from 6.5 to 4.3 within the first 7 days for 2SB and the first 14 days for 1SB, and then remained constant till the end of the experiments with T. electrotropha. The SO 4 2− concentration first increased almost 7-fold (from~350 to 2300 g/L) within the first 7 days, and then gradually till day 28 reaching 2985 mg/L for 1SB and 2545 mg/L for 2SB. Sulphide minerals were oxidised during the bioleaching of NC_01 by T. electrotropha. The total dissolved Fe increased from 0 to 0.5 mg/L within the first 7 days but remained below 0.6 mg/L throughout the duration of all the bioleaching assays. Except for 2SB, where it reached 1.5 mg/L on day 14 but went back to 0.5 g/L by the 21st day. The percentage of metal(loid)s extracted by T. electrotropha generally increased with time. By day 28, T. electrotropha leached (in %) 7-8.7 Cu, 9.6-12.4 Pb, 10.3-12.8 Zn, 23.5-24.9 Co, 2.1-4.8 As, 13.7-16.9 Cd, 0-21 K, 1.6-2.1 Sb, 3.6-5.2 Ag, and 12.7-13.9 Mn, from NC_01 via 1SB, 2SB and SMB.
The percentages of metal(loid)s solubilised by T. pacifica was comparable to that of T. electrotropha. At the end of the experiment, T. pacifica extracted (in %) between 5-8.1 Cu, 9.9-13 Pb, 9.4-12.6 Zn, 24.1-26.7 Co, 1.7-2.7 As, 13.1-16.8 Cd, 0-9.3 K, 1.1-2.2 Sb, 2.9-4.4 Ag, and 11.3-13.3 Mn from NC_01 via 1SB, 2SB and SMB. As in the bioleaching experiment with T. electrotropha, the percentage of metal(loid)s extracted by T. pacifica also generally increased with time. There was a steep increase in the total dissolved Fe in 1SB and 2SB within the first seven days, from 0 to 1.1 mg/L for 1SB and 1.6 mg/L for 2SB. This preceded a steep decrease to 0.2 mg/L by day 14 for 2SB and another increase by day 21 to 0.9 mg/L, before remaining constant till the end of the experiment. For 2SB, the Fe concentration decreased gradually after the 7th day, reaching 0.9 mg/L on the last day. The evolution in pH and SO 4 2− concentration in all the bioleaching assays were similar to that of T. electrotropha. A decrease in pH and an increase in SO 4 2− concentration during 1SB and 2SB indicated the growth and activity of T. pacifica.
Little or no sulphur oxidation occurred in the abiotic controls. The pH remained within the neutral range throughout the experiment. The SO 4 2− concentration was also constant and remained below 500 mg/L. The total dissolved Fe was either absent or very low, at less than 0.1 mg/L. In comparison to all the bioleaching assays, there was little or no metal(loid) extracted in the abiotic control assays. One exception was Mn, where the extraction efficiency of T. cyclica and T. thiocyanaticus was 7% higher than in the abiotic controls, whereas a lower equivalent extraction efficiency was recorded with T. electrotropha and T. pacifica. Figure 2 compares the maximum metal(loid) recoveries of the four halophilic sulphuroxidising bacteria. While T. cyclica mobilised substantial amounts of As (46.2%), K (31.2%), Sb (38.7%) and Ag (45.3%), the two Thioclava species both mobilised a wider range of metal(loid)s into solution. T. electrotropha and T. pacifica solubilised between 8-17% of the total Cu, Pb, Zn, Cd, K, and Mn contents of NC_01, as well as 25-27% Co. Except for Sb and Ag with good recoveries (32-39%), the maximum percentages of other elements extracted by T. thiocyanaticus were below 7%.
Little or no sulphur oxidation occurred in the abiotic controls. The pH remain within the neutral range throughout the experiment. The SO4 2− concentration was al constant and remained below 500 mg/L. The total dissolved Fe was either absent or ve low, at less than 0.1 mg/L. In comparison to all the bioleaching assays, there was little no metal(loid) extracted in the abiotic control assays. One exception was Mn, where t extraction efficiency of T. cyclica and T. thiocyanaticus was 7% higher than in the abio controls, whereas a lower equivalent extraction efficiency was recorded with T. el trotropha and T. pacifica. Figure 2 compares the maximum metal(loid) recoveries of the four halophilic su phur-oxidising bacteria. While T. cyclica mobilised substantial amounts of As (46.2%), (31.2%), Sb (38.7%) and Ag (45.3%), the two Thioclava species both mobilised a wider ran of metal(loid)s into solution. T. electrotropha and T. pacifica solubilised between 8-17% the total Cu, Pb, Zn, Cd, K, and Mn contents of NC_01, as well as 25-27% Co. Except f Sb and Ag with good recoveries (32-39%), the maximum percentages of other elemen extracted by T. thiocyanaticus were below 7%.

Bioleaching Optimisation Results
Due to the bioleaching potential observed for T. electrotropha and T. pacifica, three o timisation processes were attempted to improve their extraction efficiencies, and a four study was conducted with a consortium of both microorganisms.

Results of the Bioleaching of NC_01 by the Thioclava Consortium
The concentrations and percentages of metal(loid)s extracted from NC_01 by the T oclava consortium containing T. electrotropha and T. pacifica are presented in Tables S5 an A5; while the evolution of pH, total dissolved Fe and SO4 2− concentration are presented Figure A1. Compared to the non-optimised bioleaching experiments with T. electrotrop and T. pacifica, there was no major improvement in the extractions of Cu, Zn, As, Sb, A and Mn. However, the extractions (in %) of Pb increased slightly from 13.7 ± 0.8 to 18.6 0.8, and K from 9.2 ± 0.4 to 19.6 ± 1.3 via SMB.

Bioleaching Optimisation Results
Due to the bioleaching potential observed for T. electrotropha and T. pacifica, three optimisation processes were attempted to improve their extraction efficiencies, and a fourth study was conducted with a consortium of both microorganisms.

Results of the Bioleaching of NC_01 by the Thioclava Consortium
The concentrations and percentages of metal(loid)s extracted from NC_01 by the Thioclava consortium containing T. electrotropha and T. pacifica are presented in Table S5 and A5; while the evolution of pH, total dissolved Fe and SO 4 2− concentration are presented in Figure A1. Compared to the non-optimised bioleaching experiments with T. electrotropha and T. pacifica, there was no major improvement in the extractions of Cu, Zn, As, Sb, Ag and Mn. However, the extractions (in %) of Pb increased slightly from 13.7 ± 0.8 to 18.6 ± 0.8, and K from 9.2 ± 0.4 to 19.6 ± 1.3 via SMB. Table S6 and Table A6 present the concentrations and percentages of metal(loid)s extracted from NC_01 by T. electrotropha and T. pacifica with the serial addition of NC_01 and partial nutrient replacement. Whereas the evolution of pH, total dissolved Fe and SO 4 2− concentration are presented in Figure A2. The results showed that the percentages of the extracted Pb, Co, As, K, Ag and Mn increased slightly in comparison to the non-optimised experiments. Extractions (in %) of Co increased from 25.1 ± 0.5 to 31.1 ± 0.1, As from 4.8 ± 0.7 to 7.1 ± 0.3, K from 9.2 ± 0.4 to 19 ± 0, Ag from 5.3 ± 1 to 11.6 ± 1.2 and Mn from 14.3 ± 0.2 to 18.4 ± 0.9 via 1SB and/or 2SB. Table A7 presents the concentrations and percentages of metal(loid)s extracted from NC_01 at high temperature and pressure using the spent medium of T. electrotropha and T. pacifica. While the pH and SO 4 2− concentration before and after high temperature and pressure leaching are presented in Table A8. Only the extractions of Ag and Mn improved when NC_01 was leached at high temperature and pressure by the spent media of the two Thioclava spp. Compared to the non-optimised experiments, the recovery (%) of Ag increased from 5.3 ± 1 to 10 ± 0.9 and 11.4 ± 0.3, while that of Mn increased from 14.3 ± 0.2 to 23.2 ± 0 and 24.8 ± 0.1, at high temperature and pressure, by T. electrotropha and T. pacifica, respectively. Additionally, 2.3% ± 0.1 In was extracted by the spent medium of T. pacifica.

Results of the Microwave-Assisted Bioleaching of NC_01
XRD analysis revealed that some minerals in NC_01 (e.g., pyrite) decreased after microwave roasting, while some new minerals (e.g., hematite) were formed. The diffraction intensity peaks of some minerals disappeared or decreased in the XRD diffractograms ( Figure 3) while new peaks were formed, indicating that the initial minerals were converted into other minerals. SO4 2− concentration are presented in Figure A2. The results showed that the perce of the extracted Pb, Co, As, K, Ag and Mn increased slightly in comparison to th optimised experiments. Extractions (in %) of Co increased from 25.1 ± 0.5 to 31.1 ± from 4.8 ± 0.7 to 7.1 ± 0.3, K from 9.2 ± 0.4 to 19 ± 0, Ag from 5.3 ± 1 to 11.6 ± 1.2 a from 14.3 ± 0.2 to 18.4 ± 0.9 via 1SB and/or 2SB. Table A7 presents the concentrations and percentages of metal(loid)s extracte NC_01 at high temperature and pressure using the spent medium of T. electrotrop T. pacifica. While the pH and SO4 2− concentration before and after high temperatu pressure leaching are presented in Table A8. Only the extractions of Ag and Mn imp when NC_01 was leached at high temperature and pressure by the spent media of t Thioclava spp. Compared to the non-optimised experiments, the recovery (%) of creased from 5.3 ± 1 to 10 ± 0.9 and 11.4 ± 0.3, while that of Mn increased from 14.3 ± 23.2 ± 0 and 24.8 ± 0.1, at high temperature and pressure, by T. electrotropha and T. p respectively. Additionally, 2.3% ± 0.1 In was extracted by the spent medium of T. p

Results of the Microwave-Assisted Bioleaching of NC_01
XRD analysis revealed that some minerals in NC_01 (e.g., pyrite) decreased af crowave roasting, while some new minerals (e.g., hematite) were formed. The diffr intensity peaks of some minerals disappeared or decreased in the XRD diffracto ( Figure 3) while new peaks were formed, indicating that the initial minerals wer verted into other minerals. Tables S7 and A9 present the concentrations and percentages of metal(loid tracted from microwave-roasted NC_01 at 400, 500, and 600 °C by T. electrotropha. same time, the evolution of pH, total dissolved Fe and SO4 2− concentration are pre in Figure A3. The extractions of most metal(loid)s from NC_01 improved greatl microwave roasting. However, this improvement was not specific to the bioleach says, but was also observed for the abiotic controls (with deionised water). For ex  Table S7 and A9 present the concentrations and percentages of metal(loid)s extracted from microwave-roasted NC_01 at 400, 500, and 600 • C by T. electrotropha. At the same time, the evolution of pH, total dissolved Fe and SO 4 2− concentration are presented in Figure A3. The extractions of most metal(loid)s from NC_01 improved greatly after microwave roasting. However, this improvement was not specific to the bioleaching assays, but was also observed for the abiotic controls (with deionised water). For example, 46.2 and 49.3% Cu were extracted from the 400 • C-microwave-roasted NC_01 by 1SB and 2SB, respectively, by day 7. While 52% Cu was extracted by deionised water (abiotic control) on the same day. Nevertheless, the extractions of Pb, Ag and In were not comparable to that of the abiotic controls. Compared to the non-optimised experiments, the extraction efficiencies of Pb, Ag and In from microwave-roasted NC_01 increased from (%) 13.7 ± 0.8 to 45.7 ± 0.3, 5.3 ± 1 to 36 ± 2.6 and 0 to 27.4 ± 0.5, respectively. Additionally, T. electrotropha recovered greater amounts of Pb, Ag and In from the NC_01 samples roasted at 400 and 500 • C than the samples roasted at 600 • C. For example, the maximum percentage Pb extracted from NC_01 roasted at 400, 500, and 600 • C were 45.7, 41 and 28.8%, respectively.
Contents of the microwave-roasted NC_01 were observed to be toxic to the growth of T. electrotropha and therefore affected the bioleaching capacity of T. electrotropha. Unlike in the 2SB experiments, little or no sulphur oxidation occurred in the 1SB experiments. Instead, there was an increase in pH and little or no increase in SO 4 2− concentrations within the first 7 days. The pH in the 1SB experiments increased within the first 7 days from around 4 and 5 on day 0 to 8 on day 7. In contrast, there was a marked decrease in pH and an exponential increase in SO 4 2− concentrations in all 2SB experiments within the first 7 days. The pH in the 2SB experiments decreased from 5.6 on day 0 to between 3 and 4 on day 7. While SO 4 2− concentrations increased 24-fold from its initial concentration (136 mg/L) on day 0 to over 3300 mg/L by day 7. Additionally, most of the elements, especially Pb and Ag, were only recovered via the 2SB, and not 1SB.
As seen in Figure 4, microwave-assisted bioleaching of NC_01 by T. electrotropha was the most efficient process optimization method.
46.2 and 49.3% Cu were extracted from the 400 °C-microwave-roasted NC_01 by 1SB and 2SB, respectively, by day 7. While 52% Cu was extracted by deionised water (abiotic con trol) on the same day. Nevertheless, the extractions of Pb, Ag and In were not comparabl to that of the abiotic controls. Compared to the non-optimised experiments, the extractio efficiencies of Pb, Ag and In from microwave-roasted NC_01 increased from (%) 13.7 ± 0. to 45.7 ± 0.3, 5.3 ± 1 to 36 ± 2.6 and 0 to 27.4 ± 0.5, respectively. Additionally, T. electrotroph recovered greater amounts of Pb, Ag and In from the NC_01 samples roasted at 400 and 500 °C than the samples roasted at 600 °C. For example, the maximum percentage Pb ex tracted from NC_01 roasted at 400, 500, and 600 °C were 45.7, 41 and 28.8%, respectively Contents of the microwave-roasted NC_01 were observed to be toxic to the growt of T. electrotropha and therefore affected the bioleaching capacity of T. electrotropha. Unlik in the 2SB experiments, little or no sulphur oxidation occurred in the 1SB experiments Instead, there was an increase in pH and little or no increase in SO4 2− concentrations withi the first 7 days. The pH in the 1SB experiments increased within the first 7 days from around 4 and 5 on day 0 to 8 on day 7. In contrast, there was a marked decrease in pH and an exponential increase in SO4 2− concentrations in all 2SB experiments within the first days. The pH in the 2SB experiments decreased from 5.6 on day 0 to between 3 and 4 o day 7. While SO4 2− concentrations increased 24-fold from its initial concentration (13 mg/L) on day 0 to over 3300 mg/L by day 7. Additionally, most of the elements, especiall Pb and Ag, were only recovered via the 2SB, and not 1SB.
As seen in Figure 4, microwave-assisted bioleaching of NC_01 by T. electrotropha wa the most efficient process optimization method.

Mineralogical Analysis of NC_01 and the Bioleached Residues by T. electrotropha and T. pacifica
Results from the SEM/MLA-GXMAP showed that the waste rock was mainly com posed of silicate and sulphide minerals. NC_01 contained (wt%) 38 (Table A10). For example, 91.2% of the total Cu content of NC_0 was hosted by chalcopyrite. Surprisingly, Pb was mainly distributed between oxide

Mineralogical Analysis of NC_01 and the Bioleached Residues by T. electrotropha and T. pacifica
Results from the SEM/MLA-GXMAP showed that the waste rock was mainly composed of silicate and sulphide minerals. NC_01 contained (wt%) 38  The bioleached residues from the bioleaching experiments with the two promising bacteria (Thioclava spp.) were analysed for changes in mineral compositions. Table A11 compares the mineralogy of NC_01 before and after bioleaching by T. electrotropha and T. pacifica. Very slight (and sometimes inconsistent) changes were detected in the mineral compositions of NC_01 after bioleaching. Pyrite decreased from 15.5 ± 0.3 wt% to 14.3 ± 0.3 wt% after bioleaching by T. pacifica, but remained relatively the same (15.8 ± 0.4 wt%) after bioleaching by T. electrotropha. The total concentration of the other sulphide minerals was comparatively the same, from 1.4 ± 0.1 wt% before bioleaching, to 1.6 ± 0.5 and 1.3 ± 0 wt% after bioleaching by T. electrotropha and T. pacifica, respectively. The total concentration of oxide minerals increased from 1.3 ± 0.2 wt% to 1.9 ± 0 wt% and 1.7 ± 0.1 wt% after bioleaching by T. electrotropha and T. pacifica, respectively. For the sulphate minerals, gypsum (0.4 wt%) in NC_01 disappeared after bioleaching, while small quantities of jarosite (~0.1 wt%) were formed in the bioleached residues. The wt% of other minerals (groups) remained approximately the same after bioleaching.

Discussion
The aim of this study was to investigate the capacity of four halophilic sulphuroxidising bacteria (Thiomicrospira cyclica, Thiohalobacter thiocyanaticus, Thioclava electrotropha and Thioclava pacifica) to mobilise metal(loid)s into solution from Neves Corvo waste rock.
In comparison to all halophilic organisms screened for bioleaching, T. thiocyanaticus seemed to be the least capable of bioleaching. This is not surprising, as bacterial growth and activity were quite low in the presence of NC_01. Sulphur oxidation was limited, with little variation in sulphate concentration and an increase in solution pH, throughout the experiment. This could be due to its low tolerance to the metal(loid)s contents of NC_01. The remaining three organisms (T. cyclica, T. electrotropha and T. pacifica) displayed an impressive level of tolerance to the metal(loid) content of NC_01. This was indicated by their ability to oxidize sulphur in the presence of NC_01, assessed by the increased sulphate concentration and decreased solution pH. Though T. cyclica mobilised substantial amounts of As (46.2%), K (31.2%), Sb (38.7%) and Ag (45.3%), the two Thioclava species were the most promising organisms for bioleaching in this study. This is because they both mobilised a wider range of metal(loid)s into solution, as between 8-30% of the metal(loid)s analysed were recovered by T. electrotropha and T. pacifica. In addition, the two Thioclava species were the only organisms that seemed to utilize the sulphides present in NC_01. The amount of sulphate produced in the bioleaching media with the two Thioclava species was~43% higher than that produced by the bacterial pure cultures without NC_01 (data not shown), using the same nutrient medium and for the same duration of time. However, the pure bacterial cultures (without NC_01) of the other organisms (T. cyclica and T. thiocyanaticus), produced comparable or higher (in the case of T. thiocyanaticus) amounts of sulphate, in comparison to their bioleaching assays. This suggests that T. cyclica and T. thiocyanaticus may have only utilized sulphur compounds present in their nutrient media, and not sulphur in the waste rock. The toxic effect of NC_01 contents on T. thiocyanaticus reduced its ability to grow and oxidise sulphur, hence the lower sulphate concentrations, compared to its pure cultures where growth and the ability to oxidise sulphur were not hindered.
The rapid changes in Fe concentrations (Figure 1k) of the bioleaching experiment with T. pacifica are not unusual and have been recorded by several studies [29,30]. As Fe is solubilized from the material, it is also rapidly consumed in the synthesis of intermediate and secondary minerals, such as schwertmannite or jarosite. At any point in time, the concentration of dissolved Fe detected is the balance between mobilized Fe and consumed Fe. Therefore, higher Fe concentrations indicate that the mobilization of Fe dominated at that stage. While a lower Fe concentration indicates that the consumption of Fe dominated at that stage. To confirm that the Thioclava spp. were indeed promising for bioleaching, the bioleaching experiments with T. electrotropha and T. pacifica were repeated and similar results were obtained (Tables S8 and S9). For example, the maximum Cu recovery by T. electrotropha and T. pacifica in the first experiments were 8.7% and 8.1%, respectively. While in the second experiment, the maximum Cu recovery by T. electrotropha and T. pacifica were 8.5% and 13.1%, respectively.
The three different bioleaching assays (1SB, 2SB and SMB) helped to assess if the contents of NC_01 were toxic to the growth and bioleaching capacities of the organisms. In 2SB, biomass growth was initially separated from metal(loid) leaching by attaining the maximum cell density in the bioleaching medium, before exposing the microbe to the metal(loid)-containing solid material. This strategy could reduce the toxic effects of the solid material on the microbe, as a high biomass concentration increases tolerance to metal toxicity [31,32]. In this study, 2SB was not necessary for the bioleaching experiments, as no major difference was observed in the growth, activities or metal(loid) recovery levels of 1SB and 2SB for all bacteria. The 2SB was, however, very important for the microwave-assisted bioleaching experiment. After microwave roasting, NC_01 became quite toxic to the growth and activity of T. electrotropha. This was evident by the lack of sulphur oxidation and hence, little or no production of sulphuric acid by the microbe in 1SB. This was in addition to the consistent increase in pH and low metal(loid) recovery during the 1SB bioleaching experiments.
Microwave roasting generally transforms sulphide minerals into their more soluble sulphate and/or oxide forms, thereby increasing the extractability of target metals. After microwave roasting in this study, a high concentration of metal(loid)s in NC_01 became highly soluble, and therefore, more toxic to the growth of the bacteria. This was why little or no bacterial growth and activity were detected in 1SB. On the other hand, allowing the bacteria to grow up to the exponential stage before adding the microwave-roasted NC_01, increased the microbe's tolerance to metal toxicity in 2SB. Subsequently, the oxidation of thiosulphate in the nutrient medium by the bacteria led to an increased sulphate concentration, reduced pH and finally, increased metal(loid) dissolution. The XRD results showed that most of the pyrite present in the waste rock was oxidised to hematite (Fe 2 O 3 ) after microwave roasting at 400, 500 and 600 • C. The amount of pyrite decreased as the roasting temperatures increased, while the amount of hematite increased as the roasting temperatures increased (Table A12). Tailings originating from the same mine as the waste rock (NC_01) in this study were previously roasted in a microwave and characterised [33]. The results revealed that most of the pyrite present in the tailings was directly oxidised to magnetite (Fe 3 O 4 ) and finally to hematite (Fe 2 O 3 ) after microwave roasting at 400-600 • C. There was also a small amount of ferric sulphate (Fe 2 (SO4) 3 ), which decreased as the roasting temperatures increased. The authors concluded that the strong microwave absorption by pyrite and other sulphide minerals promoted the rapid and selective heating of these mineral phases, quickly decomposing the formed sulphates and finally resulting in a fast sulphide-to-oxide mineral transition [33]. Microwave roasting is a fast, non-contact, selective and energy-efficient heating technology [34], and it successfully improved the bioleaching rates of Pb, Ag and In by T. electrotropha in this study.
Though most sulphides were converted to their sulphate and oxide forms by microwave roasting, XRD and XRF (Table A13) revealed that pyrite and sulphur still remained in the roasted samples. Therefore, the improved bioleaching of elements from the microwave-roasted samples may have been as a result of sulphide mineral oxidation, to an extent. In addition, the bacteria produced sulphuric acid which helped to leach elements (Pb, Ag and In) that could not be leached by water (the abiotic control). The bacteria may have also synthesized some metabolites such as amino acids that could bind these elements. For example, methionine has been shown to bind strongly to Ag [35]. The thiosulphate present in the nutrient medium may have also been responsible for the leaching of Ag, as thiosulphate can effectively leach precious metals [36]. Complexation by organic acids can be ruled out, as no organic acids were detected in T. electrotropha cultures by high-performance liquid chromatography (data not shown).
The bioleaching efficiencies for metals with economic importance (e.g., Zn and Co) by the four organisms in this study were quite low. Especially when compared to the conventional (non-halotolerant) bioleaching acidophilic consortium, which recovered over 70% of the total Zn, Co, and In contents of NC_01 and 32-55% Mn and Cu, within 21 days [37]. The bioleaching rates of the two Thioclava spp. may be improved when used in combination with compatible halophilic Fe-oxidising bacteria such as Mariprofundus ferrooxydans PV-1, a neutrophilic bacterium isolated from Fe-rich microbial mats associated with hydrothermal venting at a submarine volcano, Loihi Seamount [38]. The high amount of acid-insoluble sulphide minerals (e.g., pyrite) in NC_01 necessitates the presence of Fe-oxidising organisms in the bioleaching medium, for the oxidation of Fe(II) to Fe(III). This is because Fe(III) is a more potent oxidant of acid-insoluble sulphide minerals than protons (H + ) [39], as illustrated in the thiosulphate pathway of sulphide mineral oxidation (Equations (1)-(3)) [40]. Nevertheless, the two Thioclava spp. may be more effective in bioleaching of materials with more acid-soluble minerals, such as galena or sphalerite, as the metal-sulphur bonds can be disrupted by protons [41].
The ability of the present halophilic sulphur-oxidising bacteria to oxidise mineral sulphides and mobilise metal(loid)s from mine waste into solution should raise environmental concerns. This is because, similar neutrophilic/alkaliphilic organisms with the same functions (e.g., sulphur oxidation) have been shown to be present in sulphidic mine waste dumps. For example, a high abundance of neutrophilic (halophilic) sulphur-oxidising organisms from the Halothiobacillus and Thiomicrospira genera have recently been isolated from the Neves Corvo ore and fresh tailings samples [42]. These organisms (including other types of bacteria isolated) were able to mobilise As (96%), Cd (58%), Cu (24%) and Zn (56%) from tailings obtained from an abandoned tailings dump in Barroca Grande [42]. Little is known about the activities and roles of these neutrophilic/alkaliphilic sulphuroxidising organisms in the development of acid mine drainage (AMD) in mining habitats. Yet, they are very relevant in the biogeochemistry of sulphidic ores and mine wastes. Some studies suggest that such neutrophilic sulphur-oxidising bacteria are responsible for the first series of oxidation processes that reduce the pH of fresh sulphidic mine habitats from alkaline (or circumneutral) to acidic values, before the prevalence of acidophilic ironand/or sulphur-oxidizing prokaryotes accelerating the oxidation process further at pH <3 or in AMD environments [43,44]. The bioleaching experiments presented here support these observations, indicating that the acidification of sulphidic mine dumps (e.g., tailings and waste rock) and the solubilization of toxic metal(loid)s are first-order effects of the actions of neutrophilic (halophilic) sulphur-oxidizing organisms such as the ones used in this study. Neutrophilic sulphur-oxidizing organisms are as important as acidophilic ironand/or sulphur-oxidizing organisms, as they are all strong contributors to the long-term environmental threat of acid generation and metal(loid) solubilization in sulphidic mine habitats.
The higher sulphate productions of T. electrotropha and T. pacifica in the presence of NC_01 during bioleaching, compared to the pure cultures under the same conditions, suggest that the two organisms were able to utilise or oxidise the mineral sulphides in NC_01. Therefore, a decrease in the concentrations of sulphide minerals in NC_01 was expected after bioleaching. However, no major consistent changes were detected in the mineral compositions of NC_01 before or after bioleaching by the MLA. This could be due to the very low concentrations of non-pyrite and acid-soluble sulphide minerals in NC_01 (≤0.6 wt% each) and also the limited accuracy level of the MLA technique for low concentrations of minerals. This suggests that, though the two Thioclava spp. might have changed the mineralogy of NC_01 during bioleaching, these changes occurred at minuscule levels due to the very low concentrations of several minerals and were therefore too little to be detected by the MLA with a high degree of confidence. In addition, the Neves Corvo waste rock samples were quite heterogenous, which could have contributed to the variations or inconsistent data obtained for some minerals.

Conclusions
This study demonstrated the ability of four neutrophilic and alkaliphilic sulphuroxidising bacteria to mobilise metal(loid)s into solution from the Neves Corvo sulphidic mine waste rock, in the presence of chloride ions. The low bioleaching efficiencies of the organisms indicate limited industrial potential. However, the applied and suggested optimization processes for the two promising organisms (T. electrotropha and T. pacifica) may present significant benefits, such as saline water bioleaching and the biooxidation of sulphidic ores and wastes, especially in regions with fresh water scarcity. Further studies will incorporate metal(loid) recovery from leachates using selective techniques such as sulphide precipitation and solvent extraction. The methods described in this study can be applied in other studies, as the search for competent halophilic bioleaching organisms continues. We should, however, note that the new halophilic bioleaching organisms will have to function as microbial consortia rather than as individuals, which is the key to the success of conventional bioleaching organisms.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author. The data are not publicly available yet because they are yet to be placed in an online repository. Table A1. Percentages of the metal(loid)s extracted from the Neves Corvo waste rock by T. cyclica via one-step bioleaching (1SB), two-step bioleaching (2SB), spent medium bioleaching (SMB), and the abiotic control (AC). Values are the average of duplicate flasks ± standard deviations.