Soil samples from the copper mine tailings named T1 wall of "Las Luces” (25.40.13 S; 70.35.26 W), located south of the city of Taltal, II Region of Antofagasta, Chile, were collected. From the beginning the plant was designed to use seawater without desalination in all its mineral processes. It is located very close to the sea (approximately 7 km away) (https://www.cenizas.cl/faenas_oficinas/faena-taltal/). The theoretical chemical composition of these MT are described in Table 1 (Servicio Nacional de Geología y Minería SERNAGEOMIN, for its acronym in Spanish, national service of geology and mining, 2023) (https://www.sernageomin.cl/). The soil samples were chemically characterized in ALS Life Sciences Chile S.A. according to standard protocols. Briefly, to measure metals concentrations with acid digestion and subsequent analysis by inductively coupled plasma optic emission spectroscopy (ICP-OES); for Au determination, the reference methods Au-AA23 and Au-AA24 Fire Assay Fusion were carried out (Table 2). To obtain the mineralogical analyses two methods were used: (i) quantitative evaluation of minerals by scanning electron microscopy (QEMSCAN) analysis using TIMA technology; (ii) Bulk Mineralogical Analysis (BMA) method to identify the modal mineralogical composition with the occurrence of Cu, Fe, and S (Table 3). Finally, soil pH was determined according to the protocol described by Beretta et al. (2014). Subsequently, and prior to the tests, the tailings samples were sieved at 2 mm and then quartered to obtain 5 kg of sample for all the tests carried out in this study.

From each MT samples, a small amount of soil (1 g) was immersed in 10 mL of the following nutrient media prepared both in seawater and distilled water; Luria Broth (LB), Miller (ATCC medium: 1,065), 9 k Medium (ATCC medium: 2,436) and M9 minimal medium (M9) (ATCC Medium: 2,511). Then, the mixture soil with media was incubated at 20°C and 30°C for 96 h. Samples from MT T1, an inactive tailing that is at least 10 years old, were used, then two temperatures were used to select for mesophilic environmental bacteria that might be present in the samples. To isolate indigenous bacteria, a volume of 100 µL was extracted from each cultural broth and inoculated in solid cultures of the same composition with 1.5% w/v microbiological agar. After 24 h of incubation at 20°C and 30°C, the different colony forming units (CFU), obtained by the streak culture technique in solid nutrient medium, were determined and isolated. For subsequent assays, only the bacterial isolates were used. Since during the last years, seawater is progressively replacing freshwater in the mining industry because of water scarcity from continental sources (Fu et al., 2022), the bacterial isolates were incubated into a seawater-modified LB (LB-SW) at 20°C and 30°C for 72 h.

To determine the occurrence of urease activity of the bacterial isolates, the qualitative Christensen agar method (Christensen, 1946) modified (Arias et al., 2017b) was used. Briefly, a solid culture medium was prepared containing urea and seeded in depth in 2 mL Eppendorf tubes. Subsequently, the isolates were incubated at 20°C and 30°C for 48 h. Finally, a change in color from yellow to pink was considered as urease activity positive test. S. pasteurii and B. subtilis LN8B were used as positive controls for urease activity.

Total isolates from MT samples were phylogenetically identified by 16S ribosomal gene sequencing by using the universal primers F27 (5'-GAGAGTTTGATCMTGGCTCAG-3') and R1492 (5'-TACGGYTACCTTGTTACGACTT-3'). Isolates were grown in 5 mL of nutrient medium mixed at 125 rpm for 16 h at 20°C and 30°C. The cultures were then centrifuged at 4,500 g for 5 min and the cell pellet was processed using the DNeasy Power Soil Pro^® kit (QIAGEN cat. No. 47014) according to the manufacturer’s instructions. The PCR products were sequenced by the Sanger method at Macrogen Inc. (Santiago, Chile). Finally, the resulting sequences were analyzed using Chromas Pro software, assembled and compared with the GenBank nr database by using the BlastN Program. The sequences were uploaded to GenBank with the following correlatives PP379235, PP379236, PP379237, PP379238, PP379239, PP379240, PP379241, PP379242, PP379243, PP379244, PP379245, PP379246, PP379247 and PP379248.

Urease activity was quantified by phenol-hypochlorite assay (Natarajan 1995). Briefly, urease-positive isolates on Christensen agar were cultured and their urease activity was determined at 24, 48, and 72 h. 1 mL of culture was centrifuged at 8,000 ×g for 5 min, the supernatant was removed and 174 µL of phosphate buffer at pH 8 and 434 µL of 100 mM urea were added to the cell pellet that then were incubated at 37°C for 5 min. Subsequently, further 174 µL of phenol-nitroprusside and 174 µL of alkaline hypochlorite were added to the solution and the cells incubated at 37°C for 25 min. Finally, absorbance was determined at an optical density (OD) of 626 nm. A calibration curve was previously performed with ammonium chloride in the range of 50–500 µM. Urease activity (U/mL) is defined as 1 µM of hydrolyzed urea per minute in a volume of mL of culture and it was determined considering a 2:1 (v/v) ratio between ammonium chloride and hydrolyzed urea, where for each 1 µM of hydrolyzed urea 2 µM of ammonium chloride were generated. S. pasteurii and B. subtilis LN8B strains were used once again as positive controls.

For the biocementation assays, bacterial strains were previously cultured in LB-SW medium supplemented with 40% (w/v) urea and incubated at 30°C for 48 h. Then, the cultures were centrifuged at 4,500 ×g for 15 min, the supernatant was removed, and the cell pellet was re-suspended in 50 mL of fresh LB-SW medium supplemented with CaCl₂×H₂O and urea in four different conditions according to previous studies: i) 0.5M: 1M CaCl₂×H₂O: urea v/v (Chen et al., 2017); ii) 0.3M: 1M CaCl₂×H₂O: urea v/v (Meyers et al., 2011); iii) 1M: 0.8M CaCl₂×H₂O: urea v/v (Duan et al., 2021) and; iv) 0.8M: 1M CaCl₂×H₂O: urea v/v (this study). In parallel, the MT samples were dried by hat flow at 60°C for 72 h, with the aim of removing and/or decreasing the concentration of microorganisms naturally found in the samples. Finally, a biocement samples were built as described by Omoregie et al. (2017), mixing 150 g of mine tailings with 50 mL of bacterial culture in a conical mold (base diameter of 5 cm, height of 9 cm and top diameter of 8 cm), to be incubated in batch at 30°C for 20 days. Subsequently, 1 g of the formed biocement was used to prepare serial dilutions which were grown on Christensen agar plates to determine CFUs showing as well as not showing ureolytic activity. For CFUs with ureolytic activity, from 10 to 30 CFUs were randomly selected for phylogenetic identification by 16S ribosomal gene sequencing as described in Section 2.4.

To determine the wind erosion rate (g/m²h), the biocement samples were exposed to an average wind speed of 26 km/h in an open circuit subsonic wind tunnel for 45 min, considering as erosion area the sum of the upper surface and that of the face directly exposed to the wind flow. The average velocity inside the tunnel was set according to the maximum daily prevailing wind speed measured in eight points by the Cerro Moreno Antofagasta Station of the Meteorological Directorate of Chile in the Antofagasta Region, during the year 2022 (https://climatologia.meteochile.gob.cl). Subsequently, to determine the amount of mass loss, the dehydrated at 40°C for 48 h samples were weighed before and after being exposed to the wind in the wind tunnel. The wind tunnel used was designed according to the parameters described by Moreno-Garibaldi et al. (2014) (Figure 1) and validated at the Department of Mechanical Engineering of the Universidad de Antofagasta.

To determine the mechanical strength of the biocement samples, tests of 10 mm axial compression were performed by using a mechanical press (Torin, TY10003) coupled to a Dillon PAT-705 dynamometer. The pressure exerted on the surface of the samples was determined in psi.

To determine the crystalline composition of the biocement as well as mine tailings samples before biocementation, X-ray diffraction analysis (XRD) was performed using a Bruker diffractometer model D8 Advance with DIFFRAC. EVA software and semi-quantitation of the phases with DIFFRAC. TOPAS software at the Maini Unit of the Universidad Católica del Norte. Moreover, the biocement samples were cathodically gold coated and analyzed by FE-SEM scanning electron microscopy (Hitachi, model SU5000) coupled to X-ray detection (EDX) systems (Bruker model XFlash 6,130).

The soil samples from the T1 tailings wall were characterized by determining the pH, whose average value was 7.11 ± 0.16. The semi-quantitative chemical composition detected by XRD analysis showed that the mine tailings contain mostly fractions of silicates (albite 74.058% and quartz 10.307%), followed by halite, iron oxides and calcite (Table 4). The presence of a significantly high amount of amorphous crystalline fraction (45.6%) in the sample did not exclude the presence of other chemical species, which were detected by complementary QEMSCAN analyses.

Table 3 shows the predominant mineralogical species present in the MT samples which are mainly albite, ferric oxides, quartz, and clays. The presence of carbonates is also found. The ICP-OES analysis resulted in the presence of different elements, including PTEs in different concentrations: Fe (8.01%), Al (6.83%), Na (5.94%), Ca (3.42%), Ti (1.155%), Mg (1.84%), P (1,730 ppm), Cu (1,080 ppm), followed by Mn (741 ppm), V (343 ppm), Zr (256 ppm), Ba (130 ppm), Sr (94.7 ppm), etc .,… (see Table 2 for a detailed list). These results show similarities with the theoretical composition described by SERNAGEOMIN (Table 1), but this method did not detect the presence of rare earths.

Forty-six bacterial strains were isolated from mine tailings in enrichment solutions. In LB-FW, 11 isolates were obtained at 20°C and 32 isolates incubated at 30°C. Two strains were also isolated from M9 medium in FW at 20°C and one from 9K-FW at 30°C. Sepúlveda et al. (2021) isolated 22 strains from mine tailings. In addition, growth in the different culture media qualitatively demonstrates the presence of acidophilic bacteria (9K medium), chemoautotrophs (M9 medium) and, to a greater extent, heterotrophs (LB medium). Subsequently, after immersing the isolates into LB in seawater, 23 strains were selected that were able to grow under these conditions (Table 5). During the Christensen assay, only three of the isolates showed a positive urease activity (T130-01, T130-13 and T130-14) (Table 4). The rest of the bacteria did not show any change in their color, thus showing a negative urease activity.

Phylogenetic identification by 16S ribosomal RNA gene sequencing determined the presence of nine bacterial genera (Table 5), where the genus Bacillus predominates (35%), followed by Oceanobacillus (13%), Terribacillus (13%), Virgibacillus (13%), Staphylococcus (13%), Mesobacillus (9%), Priestia (4%), Kocuria (4%) and Paenibacillus (4%). The bacteria that tested positively for urease activity were Priestia megaterium strain T130-1, Paenibacillus sp. Strain T130-13 and Staphylococcus sp. Strain T130-14, globally representing 13% of the total obtained isolates. These isolates were then characterized as Gram-positive bacteria with absence of hemolytic activity on blood agar plates (data not shown).

The pH values measured during the incubation of the strains (Figure 2A) show that at 24 h the cultures reach the ranges described for the activation of the urease enzyme, which occurs between pH six and 9 (Stocks-Fischer et al., 1999). In P. megaterium bacteria, the highest activity of this enzyme in biocementation assays is achieved between pH 7.5 to 8.5 (Sun et al., 2019). During 48 and 72 h the pH values increased to a range between 9.97 and 10.54. This increase was due to the urea hydrolysis and caused a decrease in urease activity, as demonstrated by the results of Stocks-Fischer et al. (1999) in Bacillus pasteurii, with a decrease in its urease activity by 50% approximately at pH close to 10. This result is similar to that obtained in this study.

For the isolated strains, urease activity (U/mL) was found to be similar or significantly higher in all the strains compared to the reference bacteria S. pasteurii and B. subtilis LN8B (Figure 2B). P. megaterium strain T130-1 and Paenibacillus sp. Strain T130-13 showed a bell-shaped behavior profile; at 24 h they reached values ranging from 9 to 10 U/mL, whereas at 48 h, from 14 to 16 U/mL and finally at 72 h, there has been a decrease since 10 to 6 U/mL. On the other side, Staphylococcus sp. Strain T130-14 at 24 h showed an enzymatic activity of 5 U/mL, twice lower than the other two cultures, with no significant difference with the reference bacteria. Subsequently, at 48 h it increased its activity to 11 U/mL and maintained it until 72 h. This increase was significantly higher than that of the reference bacteria.

XRD analysis of samples from T1 tailing wall and biocement samples showed the presence of quartz, albite, magnetite, halite, calcite, microcline, hematite, clinochlore and hornblende crystals (Table 5); the most abundant minerals were albite and quartz. However, it should be considered that other crystalline forms may exist and their quantification may not be reliable due to the high percentage of amorphous crystalline fraction present in all the samples (between 45.6% and 62.9%) (Table 5). Regarding the presence of calcium carbonate detected by XRD, the results are in accordance with the diffraction profile for calcite (Figure 3) of the T1 tailings wall, showing a content of 0.828% before the biocementation test. Subsequently, after biocementation test conducted with the bacterial strains, the biocement assay with the highest amount of calcite was that inoculated with Paenibacillus sp. T130-13 with a fraction of 5.412%, followed by that inoculated with P. megaterium T130-1 with a fraction of 3.191% and finally that inoculated with Staphylococcus sp. T130-14 with a fraction of 1.810% (Table 5). In all the biocementation tests with the ureolytic strains, the amount of calcite was higher than that present in the untreated MT samples according to XRD analysis in Table 5, however, according to modal mineralogical composition by BMA analysis (Table 3) calcite concentration amounted to 2.18% and only strains T130–13 and T130-1 exceeded this value.

The results obtained in the wind tunnel showed that in the biocement samples the wind erosion decreased by 78%–80% compared to the control sample (82 g/m²h) (Figure 4A). The biocementation with Paenibacillus sp. T130-13 resulted in the lowest erosion rate with a soil loss of 16.371 g/m²h, whereas the biocementation with P. megaterium T130-1 and Staphylococcus sp. T130-14 resulted in similar mass losses of 16.371 g/m²h and 17.739 g/m²h. Such erosion rates were very close to those achieved testing biocement obtained with the reference strains (S. pasteurii 18.606 g/m²h and B. subtilis LN8B 15.640 g/m²h). Then, the compression results (Figure 4B) showed that the biocement with P. megaterium T130-1 (28.578 psi) and Paenibacillus sp. T130-13 (28.404 psi) required higher pressure than the control (15.427 psi), thus improving the compressive strength capacity by 85.2% and 84.1% compared to the control. It is worthy to state that at the end of biocementation tests counts of the CFUs formed (Figure 5) were performed and molecular identification carried out by analyzing their 16S ribosomal gene. The relative results confirmed the presence of bacteria originally added in each biocementation test (data not shown). It was possible to isolate a significantly higher number of CFUs with ureolytic activity in the biocement formed with Paenibacillus sp. T130-13 (51 CFU/mL ureolytic), on the contrary to the lower number of CFUs obtained in the biocement formed with S. pasteurii (4 CFU/mL ureolytic), P. megaterium T130-1 (4.3 CFU/mL ureolytic) and negative control (10.3 CFU/mL ureolytic) (Figure 5).

Particle adhesion and agglomeration were observed in all the biocement samples where S. pasteurii and the native ureolytic strains were used (Figures 6–8). Biofilm-like structures were found to be present in cracks and spaces between particles. Particles ranging from 3 to 15 µm in size forming larger agglomerations with abundant calcium, carbon and oxygen (Figure 7C) were also observed. These structures evidenced the adhesion of the particles due to the colonization of bacteria and the formation of calcium carbonate.

The MT samples were mainly composed of feldspar (albite, plagioclase, K-feldspar) and silicates. The amount of sulfurous compounds was low, only covellite and pyrite were identified. Therefore, the probability of generating acid leachate is low. Although the pH of MT was adequate for the growth of bacteria, the variety and concentration of different PTEs were responsible for a harsh environment to an abundant growth of microorganisms.

Microbial growth in the various soil media can be interpreted as an indicator of the biological functions performed by the bacteria and the recovery status of the soil. According to Sun et al. (2019) the recovery of mine tailings begins with the action of "Keystone” microorganisms that have the ability to obtain energy from inorganic sources, thus generating: (i) the modification in the medium such as the annulment of PTEs toxicity; (ii) the stabilization of pH; and (iii) the increase of the availability of carbon sources for the subsequent colonization of heterotrophic organisms. According to the above considerations, the results obtained in this work hint at the presence of heterotrophic organisms capable of obtaining energy from organic sources, by hydrolyzing urea, since this molecule has an organic origin. The low vitality of the other isolates may be due to stress generated by the high availability of carbon sources as described for chemoautotrophic bacteria, such as Acidithiobacillus ferrooxidans ATCC 23270 and/or sensitivity to pH changes (Zhou et al., 2012; Falagán et al., 2019). Another possible reason of this phenomenon could have been the osmotic stress due to the increased salinity caused by seawater that prevented the proliferation of the isolated strains (Navada et al., 2020).

Table 5 shows that the presence of ureolytic microorganisms in soils contaminated with PTEs cannot be considered a coincidence. In a metagenomic study on mine tailings, 511 ureolytic species were identified and only 464 of them could tolerate the presence of PTEs (Hu et al., 2021). Although the isolation of native ureolytic strains in mine tailings has been successfully performed in several cases, it has not been possible to isolate an abundant number of strains. This is demonstrated in studies conducted on mine tailings in Iran, where 47 out of 76 isolates showed a positive urease activity (Jalilvand et al., 2019). Other results like those obtained in this study were obtained from copper tailings with the isolation of 22 strains and from copper-nickel mining wastes with the isolation of only two strains, thus proving the difficulty in obtaining an abundant number of ureolytic bacterial strains from mine tailings (Sepúlveda et al., 2021; He et al., 2022). On the other hand, some studies reported that native species were isolated from mine tailings and used for biocementation such as Lysinibacillus fusiformis, Bacillus firmus, Variovorax boronicumulans, Stenotrophomonas rhizophila, Bacillus sp. And Staphyloccocus equorum (Yang et al., 2016; Jalilvand et al., 2019; Sepúlveda et al., 2021; He et al., 2022; Liu et al., 2023).

The ureolytic bacteria P. megaterium strain T130-1, Paenibacillus sp. Strain T130-13 and Staphylococcus sp. Strain T130-14 represents a fraction of 13% of the total isolates obtained. No biocementation processes in mine tailings have been described using Paenibacillus sp. However, in its genome, genes for the alpha (ureC), beta (ureB) and gamma (ureA) subunits of the urease enzyme complex are encoded, in addition to other subsets of accessory genes such as ureEFD and ureFDG (Hoke et al., 2021). On the other hand, in Paenibacillus mucilaginosus, calcium and calcite biomineralization studies were conducted in industrial soils (Zhan et al., 2016) and the presence of Paenibacillus has been described in soils contaminated with arsenic, antimony tailings, lignite mine wastes, abandoned lead-silver-zinc mining waste and around mine tailings with the presence of PTEs, thus demonstrating the ability of this group of microorganisms to tolerate these contaminants (Shagol et al., 2014; Xiao et al., 2019; Karuriya and Choudhary, 2022; Nosalova et al., 2022; Wang et al., 2022).

Formerly classified as Bacillus megaterium, the bacterium P. megaterium has been studied extensively for its resistance to PTEs and its contribute in the improvement of contaminated soils for plant settlement, including trials on abandoned mine tailings (Bennis et al., 2022; Lin et al., 2022; Nie et al., 2023; Pattanaik et al., 2023). Guzman-Moreno et al. (2022) found the presence of all the genes for urease synthesis (ureA, ureB, ureC, ureE, ureF and ureD), however, urease assays resulted negative, suggesting that the expression of this enzyme could be induced only in presence of urea. The result above is in accordance with studies where carbonate precipitation, improvement of concrete properties and induction of calcium carbonate precipitates in mine tailings have been achieved using this bacterium (Lian et al., 2006; Chaurasia et al., 2019; Kim and Lee, 2019; Nasser et al., 2022). Whereas, Staphyloccocus sp., is mainly known for its negative interaction with humans, being this strain responsible for various diseases caused by its infection (Bore et al., 2007). However, this genus of bacteria is widely distributed in soils contaminated with different PTEs such as Cu, Pb, Cr, Zn, As and Hg. These bacteria are supposed to play a role in soil remediation as they enhance plant settlement in contaminated areas (Domingues et al., 2020; Fashola et al., 2020; Narayanan et al., 2020; Nosalova et al., 2022; El-Imam et al., 2023; Rojas-Solis et al., 2023; Xia et al., 2023). Like the other selected bacteria, it presents in its genome all the genes for urease synthesis (ureA, ureB, ureC, ureE, ureF and ureD), but unlike P. megaterium the urease activity in this bacterium is constitutive and when induced, its urease activity increases by 50% (Oki et al., 2010). This bacterium has been found to exhibit the ability of biocementing sand soils by precipitation of calcium carbonate and also mitigating soil liquefaction (Moosazadeh et al., 2019; Kalantary et al., 2022). Specifically, in copper mine tailings, bacteria of this genus have been used for testing co-precipitation of PTEs with calcium carbonates (Sepúlveda et al., 2021).

Ureolytic bacteria isolated from copper mine tailings showed significantly higher urease activity than S. pasteurii and B. subtilis LN8B (Figures 2A, B). P. megaterium strain T130-1 and Paenibacillus sp. Strain T130-13 presented their highest activity at 48 h, followed by a declining phase until 72 h. This trend was also observed in Oki et al. (2010) for P. megaterium, thus demonstrating that these bacteria present a constitutive urease activity and, when stimulated with urea, they can increase their urease activity by 50%. In contrast, Staphylococcus sp. Strain T130-14 shows a higher and sustained urease activity of 11 U/mL up to 72 h of culture. This behavior is similar to that described by Sepúlveda et al. (2021) for S. equorum. This trend may be related to the non-constitutive expression of urease in ureolytic bacteria as demonstrated by Guzmán-Moreno et al. (2022) in B. megaterium, where the bacterium presents in its genome all the genes for urease synthesis, but in biochemical tests did not show a positive urease activity.

After the biocementation tests, XRD showed the presence of quartz, albite, magnetite, halite, calcite, microcline, hematite, clinochlore and hornblende crystals (Table 4); the most abundant structures were albite and quartz compared to the control. In all the biocementation tests with the ureolytic strains the amount of calcite was higher than those present in the untreated mine tailings, thus proving the ability to induce calcium carbonate precipitation owned by the native strains in calcite crystalline form despite the presence of PTEs and other compounds in the T1 tailings wall. Such results are in accordance with those obtained from other MICP studies, where ureolytic bacteria in gold and copper mine tailings promoted the precipitation of calcite crystals, vaterite or aragonite (Yang et al., 2016; Kim and Lee, 2019) and with those where bioimmobilization and biomineralization assays of PTEs were conducted by MICP and the calcite formation was found (Yang et al., 2016; Sepúlveda et al., 2021; Yin et al., 2021).

The biocement obtained with ureolytic bacteria showed a wind erosion rate lower than control (no bacteria) test by 78% and 80% (Figure 4A). The lowest erosion rate was obtained in the biocementation with Paenibacillus sp. Strain T130-12. These results are similar to those obtained by Zhu et al. (2020) that conducted wind erosion tests on bio-stabilized coal mine tailings with S. pasteurii and found that the wind erosion rate decreased by 58.4% (24.57 g/m²h) compared to the control test (59 g/m²h). On the other hand, in sandy soils, biostabilized with S. pasteurii under the wind speeds ranging between 35 and 55 km/h in a wind tunnel, the decrease in mass losses resulted ranging between 99.5% and 96.5% (Maleki et al., 2016).

Additionally, results from the compression tests (Figure 4B) showed that the P. megaterium T130-1 and Paenibacillus sp. T130-13 biocement required higher pressure than the control test, thus increasing the compressive strength capacity by 85.2% and 84.1% compared to the control test. Similar results were obtained with S. pasteurii (27.102 psi) and B. subtilis LN8B (28.144 psi), showing an improvement in compressive strength by 75.7% and 82.4%. Using Staphylococcus sp. T130-14 (25.277 psi), the compressive strength was increased by 63.9%, without being statistically significant. The increase in mechanical compressive strength can be interpreted as a greater capacity to withstand soil loads and pressures (Bowles, 1997). This behavior may be due to the increased compaction of the soil particles by the precipitation of calcium carbonate between the pores and internal spaces of the soil, thus increasing the cohesion of the particles. These results indicate that the biocemented material increases its capacity to resist to mass loss by wind action, as occurs in agricultural soils where it is described that a great compaction and the presence of conglomerates decrease mass loss by wind erosion (Calderón Condori, 2017).

The mechanical strength values obtained were low compared to those described in biocement using the native strain L. fusiformis that reached pressures up to 50 psi (He et al., 2022). The value above hints at the possibility of using successfully native strains for the biocementation of mine tailings and optimize the technique for tailings stabilization as described by Buikema et al. (2017) that conducted deformation resistance tests using native strains and S. pasteurii in copper-nickel tailings, thus proving improvements in soil properties of equal magnitude between both groups.

Biocement samples resulting from tests using CaCl₂×H₂O and urea in ratios of 0.5 M: 1 M (Chen et al., 2017), 0.3 M: 1 M (Meyers et al., 2011), 0.8 M: 1 M (Duan et al., 2021), were discarded because they did not maintain the shape of the mold, thus evidencing that the mine tailings were not conglomerated (data not shown). However, the samples where a 1 M: 0.8 M (calcium: urea) ratio was used, inverse to that used by Wang et al. (2022), maintained the shape of the mold, did not deform and were malleable, even though a loss of mass was found when tapped. Since the samples with the best design were those obtained by using a higher proportion of calcium with respect to urea, this could indicate that the concentration of calcium in the medium is a critical and limiting factor in the formation of biocement, in agreement with findings obtained with S. pasteurii. Tests conducted with this train at different concentrations of calcium, showed that biocementation with higher calcium supplementation (2.5 M CaCl₂) significantly increased the hardness of the surface of the biocement samples (Zúñiga-Barra et al., 2023). On a microscopic scale, the selected biocement samples showed that in both biocementations with native and reference strains, bacteria colonized cracks and micropores inside the soil particles, being possible to isolate a significantly higher number of CFUs with ureolytic activity in the biocement formed with Paenibacillus sp. T130-13 on the contrary to the lower number of CFUs obtained in the biocement formed with S. pasteurii (Figures 5, 6). Additionally, these CFUs were a further time identified by sequencing their 16S ribosomal RNA gene, confirming the identity of the ureolytic bacteria (data not shown). Also in Figure 6 it was possible to identify by FE-SEM-EDS microscopy also the presence of calcium (yellow), carbon (blue) and oxygen (green) in the cell membrane of the bacteria, potentially indicating the formation of calcium carbonate and the role of nucleation proposed by Stocks-Fischer et al. (1999) (Figures 5D, E).

Bacterial biofilms with high calcium, carbon and oxygen content were also identified in a widespread form between tailing’s particles and on their surface (Figure 6). This would indicate the ability of bacteria to colonize the soil particle despite the presence of PTEs in the MT. In relation to the considerations above, the accumulation of calcium ions on the surface of the cells (Figure 7) and in the extension of the biofilms (Figure 6), agrees with results from Ercole et al. (2007) that described the potential role of the cell membrane and bacterial exudates in concentrating calcium ions that subsequently interact with the carbonates released by the cell and then form calcium carbonate. On the other hand, carbonates increase the neutralization potential (Qureshi et al., 2016), thus if bacteria by MICP synthesize carbonates then they will increase the neutralization potential, consequently improving the chemical stability of the MT and decreasing the possibility of generating acid mine drainage (AMD), therefore, in future studies it will be necessary to include acidity potential calculations, and in situ testing, in inactive tailings.

Despite the promise of MICP for the biocementation of MT, there are still challenges and limitations to the wide use of this technology. The most critical aspect is the release of ammonia from the ureolysis process (Erdmann and Strieth, 2023), therefore it is growing the interest in finding valuable alternatives to the use of urea and/or microorganisms that metabolize urea for biocementing MT. For example, the use of microalgae such as Anabaena variabilis (Zúñiga-Barra et al., 2023; Zúñiga-Barra et al., 2024) opens new interesting perspectives to this research, by reducing negative impacts and avoid ammonia emission into the environment.