Bioweathering of metallurgical wastes: Insights into dissolution mediated by fungus Aspergillus niger

2.1. Materials

Four types of metallurgical slags originating from copper production have been selected for this study. Detailed characteristics concerning the mineralogy of these materials have been described elsewhere: historic slag (HS), lead slag (LS) (Potysz et al., 2016a), crystalline slag (CS), and amorphous slag (AS) (Derkowska et al., 2023). Generally, AS exhibits amorphous texture and it is mainly composed of silicate glass and metallic phases. Other solids revealed crystalline texture; HS consists of fayalite, silicate glass, and sulfides, LS sample is characterized by the major volumetric proportion of sulfides, whereas CS slag reveals the most complex composition due to the presence of plagioclase, pyroxene, leucite, glass, and sulfides (Figure 1). Bulk chemical composition of the samples are presented in Table 2.

Figure 1

Slags exhibiting a wide range of chemical compositions and mineralogical characteristics were deliberately chosen to capture the diversity of metallurgical residues generated by different industrial processes. The objective was not to evaluate the behavior of a single slag type, but rather to gain broader insight into the dissolution mechanisms and bioleaching performance of materials representing a wide spectrum of geochemical properties.

2.2. Experimental approach

Metallurgical wastes were subjected to leaching experiments under abiotic and biotic conditions. Bioleaching experiments were carried out with the fungus A. niger (strain number DSM 2466, available from Deutsche Sammlung von Mikroorganismen und Zellkulturen; Braunschweig, Germany). Aspergillus niger is widely used in bioleaching studies (Table 1); this microorganism secretes organic acids, including citric, gluconic, and oxalic acids, which lower the pH and promote the dissolution of metal-bearing phases, enhancing mobilization of metallic elements. In addition, A. niger is tolerant of acidic conditions and elevated metal concentrations, making it well suited for the treatment of metallurgical slags. Its physiology, ease of cultivation, and extensive characterization in the biohydrometallurgical literature make it a suitable model organism for investigating interactions of fungi with solid materials.

Slag samples were crushed using a jaw crusher and sieved to receive a particle size of 0.25–0.5 mm. GM was prepared according to supplier’s recommendation and contained per 1 l of ultrapure water: 100 g saccharose, 1.5 g NaNO₃, 0.025 g KH₂PO₄, 0.025 g MgSO₄·7H₂O, 0.025 g KCl, 1.6 g yeast extract, and pH adjusted to 4.0. Solid materials, solutions, and laboratory equipment were sterilized in an autoclave (VWR Vapor-Line Lite II P; VWR International, USA) at 121°C, and all preparatory steps were performed under a laminar flow cabinet (CJ-600P; Chemland, Poland). Slags (10 g) were placed in a polypropylene flask (250 ml capacity) with a Hepa filter plug and immersed in 100 ml of the following solutions: (i) GM with fungus (GM + F) and (ii) sterile GM. All experiments were carried out in duplicates. Batch reactors were placed in an incubator shaker (Environmental Shaker ES-20/60, SIA Biosan, Latvia) set at 100 rpm and 30°C for a period of 30 days. The choice of experiment duration was made based on the growth cycle of the A. niger and literature survey. A typical growth cycle of A. niger (based on the biomass dry weight) will have 1 day lag phase followed by 14 days of log phase and then the stationary phase till 70 days (Aung & Ting, 2005). In our literature survey, we found that several researchers investigated fungal bioleaching by A. niger over periods ranging from 26 to 35 days (Santhiya & Ting, 2006; Xu et al., 2014; Xu & Ting, 2004).

After termination of the experiments, the leachate supernatants were separated from slag particles. Leachates obtained from abiotic leaching were filtered through 0.22 μm polytetrafluoroethylene syringe filters, acidified with 2% (v/v) Suprapur 65% HNO₃ (Merck), and stored at 4°C until elemental analysis. Leachates obtained from bioleaching were divided into two aliquots. The first aliquot was filtered, acidified, and stored following the same procedure as that applied to the abiotic leachates. The second aliquot was subjected to perchloric acid digestion at 70°C for 15 min to determine the acid-digestible elemental inventory of the fungus-containing fraction.

The slag fraction used in our experiments ranged from 0.25 to 0.5 mm. This relatively coarse fraction was intentionally selected to allow efficient separation of the slag particles from the microbial biomass at the end of the experiment. Fungal growth was assessed at each sampling time by collecting a 1 ml aliquot from each sample and aseptically transferring it into fresh sterile growth medium. The inoculated tubes were incubated under the specified conditions and subsequently examined for evidence of fungal growth.

The data acquired through Inductively coupled plasma – optical emission spectroscopy (ICP-OES) were recalculated to express them as mg kg⁻¹ of the element released from each waste (Figure 2) and to reveal the extraction efficiency (expressed as a percentage % of the total Figure 3). Procedural blanks (GM without slag and GM without slag but containing biomass) were subtracted from the corresponding leachate measurements. In addition, molar ratios of elements were calculated by converting the amounts of leached elements to moles and then determining the ratio of one element to another (e.g., Ca/Al, Cu/Fe etc.). Theoretical ratios were calculated from the bulk composition of the slags and represent the elemental ratios expected under congruent dissolution. In contrast, experimental ratios were calculated from the leaching data obtained in this study and reflect the incongruent dissolution that is more likely to occur under the experimental conditions (Welch et al., 2008).

Figure 2

Figure 3

2.3. Leachate analysis

The concentrations of elements in the leachates after the experiment were measured using the ICP-OES method (Avio 200, Perkin Elmer®). The following wavelengths were applied for the analysis: 327.393 nm for Cu, 259.939 nm for Fe, 220.353 nm for Pb, 213.857 nm for Zn, and 396.153 nm for Al, 212.412 nm for Si, 285.213 nm for Mg, and 317.993 nm for Ca. Operating conditions during analysis were as follows: plasma gas (argon) flow 12 l/min; auxiliary gas (argon) flow 0.2 l/min; nebulizer gas (argon) flow 0.6 l/min; RF power 1,500 W; pump flow rate 1 ml/min; spray chamber: TFE Tracey Spray Chamber. Calibration curves were prepared from single-element ICP-certified reference materials (Merck®).

2.4. Surface analysis

Scanning Electron Microscope with energy dispersive X-ray spectroscopy (JEOL JSM-IT100, JEOL, Tokyo, Japan) has been used for analysis of the slags before and after leaching. Fresh (unweathered) slags were prepared as polished sections, whereas slags recovered after the leaching experiments were gently rinsed with sterile ultrapure water and dried in a laminar flow cabinet. The dried grains were then used for microscopic observations to preserve the dissolution features and avoid their alteration or removal during surface polishing. Microscope operated at low vacuum and in the 15–20 kV working mode.

2.5. Statistical analysis

The results from the leaching experiments were analyzed using a non-parametric ANOVA Kruskal–Wallis test, followed by the post-hoc Dunn test with a Bonferroni correction (Kruskal & Wallis, 1952). Normal distribution of data was checked using the Shapiro–Wilk test, and homogeneity of variance (for Mg and Si) by means of Levene’s test (Shapiro & Wilk, 1965). The significance level (α) in this study is 0.05. All statistical analyses were performed in Statistica 13.3 (Tibco).

3.1. Leaching

The results of the bioleaching experiments for the elements are presented in Figures 2 and 3 as graphs with values expressed in mg kg⁻¹ and %, respectively. The pH of the leachate after termination of the experiment was 4.54 (AS), 3.25 (HS), 5.42 (LS), and 4.51 (CS). The pH drop was observed only for HS, whereas a slight increase was noted for the other materials. The decrease observed for HS was likely caused by sulfide dissolution, whereas the increase in the AS- and CS-containing solutions may be related to the buffering effect of silicates. The highest pH values observed for LS were unexpected, as this material contains a high proportion of sulfides. This suggests that, in addition to phase composition, the activity of A. niger may affect the pH of the solution by altering the H⁺ balance.

Overall, biotic conditions enhanced metallic elements leaching compared with abiotic conditions (Figure 2). For Si, Fe, Mg, Ca, Cu, Zn, and Pb, the highest recoveries were generally obtained under GM + F total conditions, indicating a substantial contribution of the fungus-containing fraction to the bioleaching process. This effect was particularly pronounced for Cu, Zn, and Pb in the LS sample and for Si and Ca in the AS sample. For instance, Cu extraction in the LS sample increased from only 0.400 mg kg⁻¹ under GM and 2.20 mg kg⁻¹ under GM + F to 9.06 × 10³ mg kg⁻¹ under GM + F total. Similarly, Zn and Pb recoveries from LS reached 1.73 × 10⁴ and 1.51 × 10⁴ mg kg⁻¹, respectively, under GM + F total conditions. In the AS sample, Si and Ca extraction under GM + F total reached 4.50 × 10³ and 5.46 × 10⁴ mg kg⁻¹, respectively (Figure 2).

The positive effect of biotic treatments was also evident in the HS and CS samples. Under GM + F total conditions, Si leaching increased from 1.65 × 10³ to 3.05 × 10³ mg kg⁻¹ in HS, while Fe recovery reached up to 9.37 × 10³ mg kg⁻¹. In CS, biotic conditions also promoted leaching, although the differences between GM + F total and GM + F were generally smaller; for example, Fe extractions were 1.32 × 10³ and 1.28 × 10³ mg kg⁻¹, respectively. Despite these improvements, HS and CS generally exhibited lower extraction efficiencies than AS and LS, particularly for Cu and Pb. In contrast, abiotic conditions (GM) consistently yielded lower recoveries, although for some elements (e.g., Ca, Zn, and Pb in selected samples), GM performed slightly better than GM + F (Figure 2).

3.2. Dissolution features

The dissolution features observed on slag surfaces following the leaching experiments are shown in Figure 4. SEM-BSE micrographs reveal representative alteration features in all slag types. AS exhibits secondary surface alteration, while unaltered glass contains voids left by the dissolution of Cu sulfides (Figure 4a–b). In HS, dissolution of the interstitial glass exposed olivine crystals (Figure 4c), and localized dissolution of sulfides was observed within otherwise relatively unweathered slag (Figure 4d). LS shows surface cracking associated with alteration (Figure 4e), together with dissolution features and the formation of secondary phases (Figure 4f). CS exhibits surface dissolution, with the notable boundary between more and less altered zones (Figure 4g), as well as rounded voids indicating dissolution of Cu sulfides (Figure 4h).

Figure 4

4.1. Susceptibility of slags to biodissolution

The results obtained by using the batch incubation tests (Figures 3 and 4) were compared for different types of Cu-slags to identify better the dissolution mechanisms and phases that control the release of individual elements from slags. The results revealed a clear relationship between the dissolution of mineral phases in the slags and the observed leaching trends of major and metallic elements.

Throughout the 30 days incubation period, the slags were exposed to a certain variation in terms of incubation conditions encompassing both abiotic and biotic leaching processes taking place at the same pH (GM vs GM + F) as seen in Figures 2 and 3. Generally, biotic incubation was found to have higher impact on elements leachability as compared to abiotic conditions (Figure 2). For this reason, phases susceptibility to dissolution presented here is considered based on results from GM + F incubation variant.

One of the main differences between the leachability of metallic elements and that of major elements is the fact that metals are associated with sulfides and intermetallic phases (Cu-bearing droplets), while major elements mainly occur in silicates and glass. Consequently, the leaching tendency of individual elements may vary from one phase to another and is specific to the overall phase composition of the slag (Figure 3). An analysis of Cu leaching behavior revealed that AS and LS samples exhibited the highest susceptibility, with leaching rates reaching 25.1–29% (GM + F total) (Figure 3). However, given the entirely different phase compositions of these materials, it can be stated that amorphous texture of AS played main role in determining its susceptibility to leaching (Figure 3). On the other hand, Cu in LS was primarily incorporated into sulfides, which also serve as the main Cu carriers in HS and CS. This suggests that the volumetric dominance of sulfide phases in LS promoted their greater exposure to the leaching environment, thereby enhancing subsequent sulfide dissolution and element release, whereas in HS and CS, the presence of crystalline silicate phases, where sulfides are encapsulated impeded leaching by reducing sulfide exposure compared to LS (Figure 4). Furthermore, the highest Zn leaching was achieved for AS slag (27.3%, Figure 3) likely due to its amorphous texture, whereas CS revealed the following leaching behavior: CS (57.4%) > HS (9.1%) > LS (8.7%) (Figure 3). It must be emphasized that LS contains sphalerite (Zn-sulfide) in the highest volumetric proportion (Potysz et al., 2016) despite its leaching being lower than that of CS and HS. In the former slags, the Zn impurities in crystalline phases and glass were primary contributors to Zn release. This shows that greater susceptibility to dissolution of crystalline phases can be observed rather than Zn-bearing sulfides. This finding contradicts the observations of LS dissolution in organic acids reported by Potysz et al., 2017. They suggested that a higher concentration of biotic agents may be necessary to enhance sphalerite leaching. As reported by Nnanwube et al., (2024); Nnanwube and Onukwuli, (2020), the sphalerite leaching rate increases with higher organic acid concentrations. Furthermore, slags revealed the following tendency to Pb leaching: AS (99.3%) > LS (27.6%) > CS (11.3%) > HS (2.8%) (Figure 3). Again, the amorphous texture of AS was the primary variable determining the behavior of this material (Figure 4). In LS, intermetallic phases were the main Pb carriers, known for their susceptibility to leaching, whereas in CS and HS, the dissolution of sulfides was responsible for the primary Pb input to the solution.

The calculation of the molar ratios of the individual elements is a useful geochemical tool for deciphering which phase is being preferentially dissolved under specific weathering conditions (Figure 5). An important observation derived from the calculated molar ratios is that these values achieved for GM + F total are similar to those in a bulk sample. This result suggests that the presence of A. niger (considering its biomass) affect slags promoting its bulk dissolution. When all the elements were compared to Cu, regardless of incubation conditions and the material studied, Cu is leached less preferentially than the major elements and Zn (Figure 5). An exception from this rule was observed for AS sample for which Cu was leached preferentially. However, Cu exhibited preferential leaching over Pb. Even though slag exposure to biotic factors can contribute to Cu leaching, the major phases (glass) still show greater susceptibility to dissolution (Figure 5). Tracking the leaching of the major elements allowed to relate materials behavior to the dissolution of silicate phases. Focusing solely on Si would have made it impossible to decipher the dissolution process; therefore, a broader analysis encompassing additional elements was conducted. Calcium was released in the quantities 43.7% (AS) vs 3.31% (CS), which again suggests greater susceptibility of AS as compared to crystalline one (Figure 4). In addition, Ca (3.31%) vs Al (1.31%) in CS (Figure 3) indicates that higher susceptibility of plagioclase and/or pyroxene rather than leucite took place (Figure 5). As reported by Welch and Ullman, (1993), organic acids (such as those involved in the studied bioleaching) enhance plagioclase dissolution even by ten times compared to inorganic solutions at the same acidity. In addition the authors pointed out that Ca leaching from plagioclase reveals faster rates than that of Al and Si (Welch & Ullman, 1993), which is also in accordance with the present study (Figures 3 and 4). In HS, the main indicator for crystalline phases dissolution was Fe-bearing fayalite; in this case, the Fe release reached 2.6%, which was lower than that observed for other materials LS (35.5%) > AS (7.7%) > CS (4.0%) (Figure 3). Based on that, it is concluded that silicate phases of HS are similarly susceptible to dissolution as crystalline phases of CS, which is also confirmed by Si leaching reaching 1.7% and 1.5% for HS and CS, respectively. In addition, calculated molar ratios indicate that glass of HS is generally preferentially dissolved with respect to fayalite except for GM + F total incubation conditions, where the impact of fungus on fayalite dissolution becomes more notable (Figure 5). Interestingly, despite fayalite occurring in LS in minor volumetric proportions, its leaching was greater than that of HS, which contains fayalite as a volumetrically major phase (Figure 4). This statement is confirmed by leaching results showing Si and Fe leaching from HS, 1.7% (Si) and 2.6% (Fe) compared to leaching from LS at 15.8% (Si) and 35.5% (Fe).

Figure 5

4.2. Effect of incubation conditions on slags behavior

In this study, two incubation conditions were investigated, which included slags immersion in GM representing abiotic leaching and medium inoculated with fungus A. niger representing biotic leaching mechanism. In addition, the biomass of A. niger was mineralized to differentiate between the amount of elements released into the dissolved fraction and those associated with the microbial cells. Regardless of the slag studied, it was shown that the concentration of elements in the leachates was similar under both abiotic (GM) and biotic (GM + F) conditions, whereas notably higher concentrations of elements were determined in fungus-associated fraction (Figure 2). This statement is supported by the statistical analysis (Table 3), which demonstrates that differences in Si, Fe, Mg, Cu, and Zn concentrations in leachates (across all types of slags used) between biotic and abiotic conditions are not statistically significant. Regarding the fungus-associated fraction, notably higher Mg content is observed compared to biotic and abiotic conditions (Table 3). For the other elements, differences are more diversified. It must be emphasized that elements detected after biomass digestion may represent surface sorption, intracellular uptake, oxalate precipitation, and secondary phases formation (Javanbakht et al., 2014; Warren & Grant Ferris, 1998). The abundance of functional groups (carboxyl, amine, hydroxyl, phosphate, sulfonate) in fungal biomass may be an excellent biosorbent for metals. In our case, the living microbial cells of A. niger were present in the incubation experiment and therefore the active mechanisms of detoxification and biosorption took place and could effectively enhance metal sorption (Malik, 2004). The carboxylate and amine groups of A. niger biomass were more important biosorbing sites than the phosphate and the lipid functional groups (Kapoor & Viraraghavan, 1997). Infrared spectroscopy analysis revealed the binding sites in A. niger such as two absorption bands in the 3,690–3,615 and 2,970–2,895 cm⁻¹ regions correspond to OH and CH groups in fatty acids and bands around 1,600 cm⁻¹ and 1,048 cm⁻¹ correspond to C═O groups of amides (Gáplovská et al., 2018). The biosorption of metals such as Cu, and Zn by A. niger generally obey Langmuir and Freundlich models (Filipovic-Kovacevic et al., 2000; Mukhopadhyay et al., 2007). When the total amount of elements (GM + F total) released during the incubation experiment was considered, most elements showed higher concentrations than those observed in the dissolved fraction (GM + F), and the leaching yield was improved by factors up to 68.6 (Ca), 27.8 (Fe), 13.6 (Mg), 3.2 (Al), and 1.3 (Si) (Figure 3). The differences between GM, GM + F, and GM + F total were particularly pronounced for several elements in the HS slag (Figure 2). For Ca, the recovered amount increased from 0.2 mg kg⁻¹ in GM to 32.4 mg kg⁻¹ in GM + F and 237.4 mg kg⁻¹ in GM + F total, while for Mg the corresponding values were 322.5, 486.6, and 1,163.9 mg kg⁻¹, respectively. The most pronounced difference was observed for Cu, for which the recovery increased from 0.6 mg kg⁻¹ in GM to 2.8 mg kg⁻¹ in GM + F and 184.7 mg kg⁻¹ in GM + F total. These results demonstrate that the contribution of the fungus-associated fraction to total element recovery was strongly element-dependent. Similar, although less pronounced and less consistent, differences between the incubation conditions were observed for the other slag types, indicating that element partitioning between the dissolved and fungus-associated fractions depends on both element type and slag composition. However, it must be emphasized that element binding to biomass varied from one solid to another, demonstrating that not only the general affinity of the element to biomass but also the overall element dissolution, determines fungal capture. Among the major elements, Ca was the most intensively incorporated to biomass (factor up to 67). The ability of A. niger to capture Ca has already been reported (Su et al., 2023). Additionally, calcium oxalate particles formed on fungus hyphae have been previously demonstrated. The bioleaching process enhances element dissolution by promoting their complexation with organic acids released by the fungus into the leaching medium. Thus, once Ca is captured by fungal hyphae, the competition of this element with other metals is reduced (Xu et al., 2014). Metallic elements exhibited the following tendency to bind to biomass: Cu (factor up to 4,000), Zn (factor up to 98.5), and Pb (factor up to 80.1). Apart from Cu sorption, Zn was also reported to be effectively captured by A. niger (Hajahmadi et al., 2015). This is an important result indicating that studies dedicated to incubation with microorganisms forming dense biomass must consider not only the dissolved element fraction but also the fraction embedded in biomass to draw more reliable conclusions on phase susceptibility to bioleaching (Table 3).

4.3. Overall effect of bioleaching approach investigated for the studied slags

Bioleaching was especially efficient for the extraction of Zn from CS (57.4%), Cu from AS (29.1%) and LS (25.1%), and Pb from AS (>99%). These yields were obtained under non-optimized conditions (10% pulp density, 0.25–0.5 mm particle size, 30 days), which were deliberately chosen to reflect industrially relevant solid loadings and to facilitate biomass separation, rather than to maximize leaching efficiency. The observed differences in metal recovery between slag types are therefore primarily attributed to their mineralogical composition, with amorphous phases in AS and sulfide-rich phases in CS and LS being more susceptible to fungal-mediated dissolution. At the same time, notable leaching of major elements (e.g., Ca and Fe) occurred, which would require additional purification steps in any potential recovery process. Another important finding is the sequestration capacity of live A. niger biomass (Figure 3), as evidenced by the consistently higher concentrations of Cu, Zn, and Pb in the GM + F total fraction compared to GM + F and GM alone (Figure 3). This is consistent with previous reports on the sorption of divalent metal ions by A. niger biomass (Kapoor & Viraraghavan, 1997; Liu et al., 2006; Mukhopadhyay et al., 2011). However, direct comparison of our bioleaching efficiencies with literature data (Gomes et al., 2020; Schippers, 2017) is limited because this study was not designed to optimize bioleaching parameters. The relatively high pulp density (10% w/v) and long incubation time (30 days) suggest that a broader range of parameters should be considered in future experiments. A methodological perspective is presented in Section 5.