Acid mine drainage (AMD) samples were collected from Margherita coal mines in northeastern India. During sampling, the onsite pH and temperature were measured and recorded. The samples were then transported to the laboratory for isolation. Modified 9K media was used for the isolation of target bacterial strains [33]. The isolation and identification procedure as per Bhaskar et. al., (2019) [34]. The nucleotide sequence was submitted to NCBI Genebank to get an accession number.

Fly ash was collected from the plant next door in Karnataka, India. For bioleaching, modified 9K media without iron supplement were used, and the appropriate amount of fly ash was added to provide an iron source and make up a 100 ml volume of the mixture [35]. An isolated Acidithiobacillus ferrooxidans inoculum of 1.0×10⁷ cells/ml was added to this mixture to initiate the leaching studies [34]. All the experiments were conducted with sterile conical flasks on an incubator shaker at a temperature of 32°C (BioBee, 350 L, Semi-automatic, Orbital 50–250 rpm). For 15 days, samples were taken for analysis at regular intervals (daily). Samples were obtained during sampling, filtered using a syringe filter, and then collected for further study. X-ray diffraction investigations (Ultima III series, Rigaku, TSX system, Japan), electron dispersive spectroscopy (Hitachi Noran system 7, USA attached to the Zeiss supra 55 VP), and scanning electron microscopy (Hitachi, Japan), were used to characterize raw fly ash before being subjected to bioleaching. The potassium thiocyanate technique was used to quantify the ferric iron content using a double-beam UV-vis spectrophotometer (PerkinElmer, Lambda 35 uv-vis) [36]. The concentration of Total iron was measured by the 1, 10 phenanthroline method using a double beam UV - Vis spectrophotometer (PerkinElmer, Lambda 35 uv-vis) [37]. pH was measured using a digital pH meter (Model-edge, HANNA). Oxidation and reduction potential were measured with a Redox meter (EUTECH, −999 to +1000 mV).

For optimization, the Box-Behnken design Response Surface Methodology is used. The response surface approach was used to investigate the iron dissolution. With fifteen iterations, we identified three levels (−1, 0, +1) to serve as independent variables and iron dissolution as the dependent variable (Table 1). The variable parameters were optimized in the range of 50 rpm to 150 rpm for Shake flask speed, pH in the range of 2.5 to 3.5 and pulp density 1% to 5% with no sulfate supplement and 1 g/L to 10 g/L sulfate with sulfate supplement and iron dissolution was considered as a response. For experimentation with sulfate supplement pulp density of 2.5% was considered. The improved factors that affect the responses were evaluated using 3D contour graphs after fitting the experimental data to a regression model.

Dark reddish colonies obtained on the plate morphologically confirmed the iron oxidation by the bacteria. About twelve isolates were obtained from plated samples. Out of which three isolates exhibited growth on re-streaking and a single pure isolate was finalized based on the growth. Physical morphological studies show that those isolates obtained were gram-stained and confirmed the Gram −ve reaction. The isolate showed efficient growth at pH 2.5 and 30°C temperature on a modified 9K medium, ferrous being the energy source.

Using Nanodrop, the DNA concentrations of the two isolates were determined to be 26.4 ng/μL and 28.7 ng/μL. An amplicon measuring 900 bp was obtained during amplification with universal primer pairs 8F and 1391R. Additionally, an amplicon size of 900 bp was obtained by amplification utilizing the species-specific primers F1_Thio (Sense) and R1_Thio (Antisense). After amplified DNA underwent gene sequencing, the sequence findings were compared using NCBI BLAST software, which revealed that Acidithiobacillus ferrooxidans bacteria shared 98% of the sequences. The bacterial strain was assigned accession number OP326196, with the bacteria designated Acidithiobacillus ferrooxidans, when the isolation and sequencing uniqueness were confirmed.

Fly ash powder can be seen as round globular particles with smooth surfaces. Fig. 1 presents the SEM and EDS images of fresh fly ash particles. Peaks observed with XRD (Fig. S1) at 2θ 18.28, 37.42 and 74.34 indicates the presence of Iron Silicon Oxide (PDF: 01-089-6228), peaks 2θ at 30.28, 43.33, 53.81, 57.25, 62.87 correspond to Hematite (01-079-0007) and peaks at 2θ 90.03 and 94.99 correspond to Aluminum ruthenium (00-029-1404) [19]. Fresh fly ash particles on Fresh fly ash particles on EDS analysis were revealed to have 3.2% elemental iron content (Fig. 1). XRF analysis depicts that fresh fly ash particles have 37.02%, 0.2105%, 0.7184%, 0.089% and 0.0001% iron oxide, aluminium oxide, silicon oxide, magnesium oxide and ruthenium respectively while bioleached fly ash particles have 0.029%, 8.337%, 0.0946%, 0.247% and 0.000009% iron oxide, aluminium oxide, silicon oxide, magnesium oxide and ruthenium respectively (Table S1). About 99.9% iron dissolution was observed from XRF analysis. In bioleached fly ash, there was an increase in magnesium and aluminium oxide. This is due to the nature of the 9K medium, which may have aided in the formation of magnesium oxide and aluminium oxide. The presence of aluminium, silica and carbon was attributed to the production process. Four distinct bands on FTIR analysis, centred at around 1635, 1223, 888, and 604 cm⁻¹, are seen for fresh fly ash samples (Fig. 2). The cause of the strong broadband at 1635 and 1223 cm⁻¹ is either asymmetric stretching of Si-O-Si or Si-O-Al. Si-O-Si represents siloxane bonds between silicon atoms bridged by oxygen which are stronger and stable while Si-O-Al is aluminosilicate bonds with aluminium substituting some silicon atoms possessing high reactivity with a negative charge. In fly ash, it is observed both Si-O-Si and Si-O-Al are present with the relative concentration of aluminium content. The presence of aluminium induces more structural variability and potential reactivity to fly ash. The Si-O-Si and O-Si-O bands are located at 888 and 604 cm⁻¹, respectively. The surface −OH groups of silanol are responsible for the broad spectra of 3465 cm⁻¹. The broadness of the link (spectral peak) indicates a strong hydrogen bonding [38–39]. On bioleaching, it is observed that the peak shows a modest shift in FTIR spectra (Fig. 2.).

Model F’s score of 41.79 suggests that it is important for the bioleaching approach of dissolving iron without the use of a sulfate supplement. The probability of an F-value occurring owing to noise is merely 0.01%. If the model’s probability value is less than 0.0001, it is considered significant. An indication of how well the model fits the data is the regression coefficient, R² = 0.9817, which shows a high degree of correlation between the anticipated and observed values. Based on their value, factors that are very significant for the bioleaching of iron from the fly ash model explain over 98.17% of the variation. Only 1.83% of the total variance is not explained by the model. The ANOVA quadratic regression model demonstrated a high level of significance for iron bioleaching, as shown by a low probability (p < 0.0001) of the F-test and an insignificant lack of fit (Table 2). The lack of fit F-value of 5.41 indicates that the pure error does not significantly contribute to the lack of fit. A strong lack of fit F-value has a 6.82% probability of being caused by noise. The discrepancy between the adjusted R² values of 0.9582 and the projected R² values of 0.7598 is less than 0.2, indicating a satisfactory agreement.

The model F-value of 75.90 for the bioleaching of iron with sulfate supplement suggests that the model is important for the bioleaching method’s iron dissolution from fly ash. If the model’s probability value is less than 0.0001, it is considered significant. Regression coefficient R² = 0.9992 is a metric of model fit quality that shows a strong degree of correlation between the observed and predicted values. According to the value, variables of high relevance for the bioleaching of iron from fly ash on the sulfate supplement model account for more than 99.92% of the variance. This means that the model can only account for 1.01% of the total variance. An increase in initial density resulted in the release of iron as a substrate for Acidithiobacillus ferrooxidans, hence enhancing the recovery of iron from the fly ash [40].

For the bioleaching of iron from fly ash on sulfate supplement, the ANOVA quadratic regression model showed that the model was highly significant with a low probability (p < 0.0001) of the F-test and insignificant due to lack of fit. The lack of fit F - value of 2.43 suggests that the pure mistake has no major bearing on the lack of fit. A significant F-value for lack of fit could be the result of noise with a 20.50% likelihood. The discrepancy between the adjusted R² values of 0.9768 and the projected R² values of 0.8895 is less than 0.2, indicating a satisfactory agreement (Table 2).

The bioleaching model for iron extraction from fly ash exhibits a coefficient of variation of 4.19%, suggesting its high level of reproducibility. The CV value serves as an indicator of the trial’s exceptional accuracy and reliability. The p values of the regression coefficients reveal that the interaction and linear test variables have a substantial and statistically significant effect on the bioleaching of iron from fly ash. With a coefficient of variance of 4.43% for the bioleaching of iron from fly ash on sulfate supplement, the model can be deemed reproducible. The excellent precision and dependability of the trials are indicated by the CV value. The impacts of the independent test factors, which include linear, quadratic, and interaction effects, on the bioleaching of iron from fly ash on sulphate supplement are very significant, according to the regression coefficient’s p values [41].

In the present study, A, C, AC, A², B² and C² are significant model terms for the Bioleaching of Iron from fly ash (Table S2). Thus, statistical analysis of all the bioleaching experimental data showed that Shake flask speed, pulp density and pH had a significant effect on the bioleaching of Iron from fly ash. It is observed that shake flask speed and pulp density exerted a more linear influence which indicates that iron dissolution by the process was positively influenced by shake flask speed, pulp density and pH. A higher positive linear effect of shake flask speed and pulp density over pH in bioleaching has been reported by researchers in previous work [19–20, 42]. Again, shake flask speed has the highest positive linear influence (high co-efficient) compared to pulp density and pH due to enhanced mixing of the fly ash particles improving contact between bacterial culture and fly ash particles. This would facilitate better mass transfer and access for the bacteria to dissolve iron. Also, it should be noted that higher agitation speed increases oxygen transfer into the liquid media which is important for the growth and metabolic activity of A. ferrooxidans. The quadratic effect of the independent variables on the Bioleaching of Iron from fly ash was negative. pH has less effect than shake flask speed and pulp density in the model. The reason for this may be attributed to the adaptation of A. ferrooxidans strain in the pH range of 2.5 to 3.5 as the generation time is based on cellular activity in response to pH stress [43].

The linear regression quadratic equation of the model fit is explained by Eq. (1).

where A, C, A², B² and C² are significant model terms for the Bioleaching of Iron with sulfate supplement (Table S3). Thus, statistical analysis of all the experimental data showed that shake flask speed and sulfate addition had a significant effect on the Bioleaching of Iron with sulfate supplement. It is observed that shake flask speed and sulfate supplement exerted a more linear influence which indicates that Iron dissolution was positively influenced by shake flask speed and sulfate supplement. Again, the sulfate supplement has the highest positive linear influence (6.61) than the influence of shake flask speed (6.29) and pH (2.38). The quadratic effect of the independent variables on the Bioleaching of Iron with sulfate supplement was negative (not linear) which means as the independent variables increase, the bioleaching of iron initially increases up to a certain point (optimum level), then begins to decrease reducing the effectiveness in bioleaching.

The linear regression quadratic equation of the model fit is explained by Eq. (2).

Significant factors affecting iron dissolution are shake flask speed, pulp density, pH, and their interactions with maximum iron dissolutions of 75.5 mg/L while Bioleaching of Iron from fly ash with sulfate supplement has a comparatively broad range with sulfate supplement with the maximum iron dissolution of 94.5 mg/L (Fig. 3). Actual and predicted responses for both the Bioleaching of Iron from fly ash and the Bioleaching of Iron from fly ash with sulfate supplement are highly relevant. Fig. S2 shows the interaction plot for bioleaching without sulfate addition, revealing the synergetic effect of shake flask speed and pulp density on iron dissolution while Fig. S3 presents the interaction plot for bioleaching with sulfate supplement, demonstrating how sulfate supplementation modifies the impacts of other variables on iron extraction. These plots help in identifying the most favourable combinations of variables for maximizing iron dissolution.

The influence of independent variables on dependent variables was analyzed using a Pareto chart (Fig. S4). In the case of Bioleaching of Iron from fly ash A (Shake flask speed), C (Pulp Density), AA (Effect of Shake flask speed), CC (Effect of pulp density) and AC (Combinational effect of shake flask speed and pulp density) have the significant effects. The interaction between shake flask speed and pulp density (AC in the model) was found to be significant, indicating a synergistic effect on iron dissolution. Proper agitation and pulp density ensure an even distribution of nutrients and metabolites throughout the medium, supporting consistent microbial growth and activity. Higher shake speeds can help overcome diffusion limitations, especially at higher pulp densities, by improving the movement of leached metals away from particle surfaces [19–20]. while for Bioleaching of Iron from fly ash on sulfate supplement A (Shake flask speed), C (Concentration of sulfate), BB (Effect of pH), AA (Effect of Shake flask speed) and CC (Effect of Sulfate supplement) have significant effects. Sulfate supplement enhances bacterial growth promoting their metabolic activity by providing essential nutrients. Sulfate supplementation can provide a buffering effect, maintaining a stable pH conducive to bioleaching over extended periods. Sulfate ions can modify the surface properties of fly ash particles, potentially increasing their susceptibility to bacterial attack. Low pH prevents iron hydroxide precipitation, keeping iron in solution for further processing, and modifying fly ash particles surface properties [44–45].

From the Fig. 8 contour plot, it is observed that Bioleaching of Iron from fly ash works in a narrow range of shake flask speed and pulp density (Fig. 4(a)). It is to be noted that pulp density is a critical factor without sulfate, while sulfate supplement contour plot is shown to have a broader effect as sulfate concentration becomes more important (Fig. 4(b)). Fig. 5 shows a 3D response surface plot for the Bioleaching of Iron from fly ash and the Bioleaching of Iron from fly ash with a sulfate supplement. Without sulfate supplement, the curved surface of the 3D plot indicates a non-linear relationship between shake flask speed and pulp density on iron dissolution. While with sulfate supplement, the curved surface is a broader optimal region compared to the without sulfate supplement condition indicating higher iron dissolution rate at higher shake flask speed and sulfate supplement. By the contour plots, it is observed that shake flask speed remains an important factor in both cases and pulp density is a crucial factor without sulfate addition while sulfate concentration is important in the latter case. Fig. S5 presents iron dissolution from fly ash by bioleaching method without and with sulfate supplement in which pulp density has a significant effect in the case of bioleaching with no sulfate supplement and concentration of sulfate has a significant effect in the case of bioleaching with sulfate supplement. The overview of Bioleaching of Iron from Fly ash without and with Sulfate addition is presented in Table 3. Table S4 depicts the comparison of iron leaching from fly ash with previous studies. Iron leaching achieved in the present study is 62.72% which is much higher than previous bioleaching attempts (45.90%) as achieved by Bhaskar and team [19]. The acid digestion method is observed to be less effective with 30–93% iron leaching efficiency compared to the present bioleaching method with sulfate supplement [46–47]. Carbon thermal leaching reported by Wang and team (2017) achieved 96.94% iron leaching [48]. The sulphate supplement on bioleaching of iron from fly ash in this study notably enhanced iron leaching from 62.72% to 99%, showcasing the efficacy of this method.