Optimization of biological sulfuric acid generation by Acidithiobacillus thiooxidans in a stirred-tank reactor system

Abstract

Abstract

Conventional sulfuric acid production is tied heavily to the fossil fuel industry, where it is typically obtained from the desulfurization of fuels. Here, the biological production of sulfuric acid by Acidithiobacillus thiooxidans in a reactor setup was optimized using design of experiments methodology. The effects of initial sulfur dosage, phosphate (PO₄³⁻) concentration, and inoculum size were evaluated with respect to sulfate (SO₄²⁻) concentration, optical density, and pH. A significant effect was observed for sulfur dosage; the highest sulfur supplementation, 35 g L⁻¹, yielded the highest SO₄²⁻ concentration and lowest pH after 14 days of cultivation. A mean sulfur-to-SO₄²⁻ conversion rate of 81 ± 9% was determined across all trials conducted with 35 g L^− 1 of sulfur. Among all trials, a maximum SO₄²⁻ molarity of 1.08 M was achieved, with a corresponding pH of 0.18. No significant effects were determined for PO₄³⁻ and inoculum volume. The feasibility of using an industrial by-product as a sulfur source was demonstrated, in addition to establishing a strong correlation between the measured SO₄²⁻ concentration and conductivity of the biological lixiviant.

Key points

• Sulfur dosage correlated positively with SO₄²⁻ concentration.

• DoE revealed no significant effect for PO₄³⁻ concentration and inoculum size.

• Strong correlation between SO₄²⁻ concentration and conductivity was established.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1007/s00253-026-13866-2.

Introduction

The recycling of secondary resources relies largely on energy-intensive pyrometallurgical and reagent-heavy hydrometallurgical technologies. Sulfuric acid is a common inorganic acid used in various industrial fields; however, its generation is highly dependent on oil and gas extraction. Additionally, a variety of green technologies require sulfur in the form of sulfuric acid, further contributing to its increasing demand (Maslin et al. 2022). Microorganisms such as Acidithiobacillus thiooxidans belong to the group of sulfur-oxidizing bacteria (SOB) and are classified as acidophilic, mesophilic, autotrophic, aerobic (facultatively anaerobic), and gram-negative microorganisms that play a major role in the geochemical cycling of sulfur. Through the oxidation of reduced inorganic sulfur compounds (RISCs), electrons are generated, which then enable A. thiooxidans to derive its energy through respiratory processes (Ibáñez et al. 2023). It is commonly found in sulfur-rich environments such as sulfur springs, caves, and in areas affected by acid mine drainage. Acidithiobacillus thiooxidans grows optimally at 28–30 °C; however, is able to survive a broad temperature range from 10 to 37 °C (R. Wang et al. 2019). In terms of acid tolerance, A. thiooxidans has been reported to thrive at pH values < 0.5 (Feng et al. 2015; Ramirez et al. 2009).

Acidithiobacillus thiooxidans is commonly associated with geological processes occurring naturally (Shen et al. 2022) or within a biotechnological context aimed at the extraction of metals and the removal of sulfides (Gómez Ramírez et al. 2020; Hocheng et al. 2014; Y. J. Lee et al. 2020; Ramirez et al. 2009; Yang et al. 2019). Previous studies have investigated the potential of using direct bioleaching for the extraction of valuable metals from waste products, which allows direct contact between microorganisms and waste material (Bayat et al. 2009; Gomes et al. 2018; J. Wang et al. 2009). Nevertheless, current literature favors the indirect (non-contact) bioleaching mechanism, as bacterial metabolites such as protons and ferric iron induce metal extraction (Gu et al. 2018; Xin et al. 2009). Separating the generation of biogenic acid from the subsequent leaching processes may result in higher acid strength along with improved process control (Kremser et al. 2022). Furthermore, higher pulp densities can be applied as the microorganisms are not directly exposed to heavy metals, thereby avoiding limitations due to bacterial inhibition and improving economic feasibility. Low pulp densities are associated with high energy consumption for reactor stirring and temperature control as well as high reagent costs, increasing the environmental footprint while decreasing the economic competitiveness (Alipanah et al. 2023).

The generation of biological acids using acidophilic microorganisms in a reactor has been studied previously (Kremser et al. 2022; Pakostova & Johnson 2019). Commonly reported sulfur dosages for bioleaching experiments range around 10 g L⁻¹ (Adetunji et al. 2023; Gómez Ramírez et al. 2020; Lin et al. 2010; Van Yken et al. 2023); in contrast, Kremser et al. (2022) tested dosages up to 60 gL⁻¹ sulfur and reported a maximum of 64% sulfur-to-sulfate (SO₄²)⁻ conversion after 3 weeks of cultivation. In Table 1 a series of previously published experiments are listed where A. thiooxidans has been studied specifically in terms of its sulfur-oxidizing potential. Except for the experiments conducted by Kremser et al., all measured SO₄²⁻ concentrations were found to be below 1 M; however, Kremser et al. required 28 days to achieve said concentration. In order for biohydrometallurgical processes to classify as economically viable, the process duration and the solid to liquid ratio during the actual leaching process are of crucial importance. Longer process durations require higher energy requirements for aeration and stirring, which in turn has the largest impact on all environmental impact categories investigated within a life cycle assessment (LCA) (Thompson et al. 2018). On the other hand, higher acid molarity allows for the implementation of higher solid to liquid ratios, resulting in a higher throughput of material to be processed within a given time frame, thus directly impacting the energy requirements to process a ton of material.

Table 1.

List of published experiments using A. thiooxidans to oxidize elemental sulfur

Strain	Sulfur [g L−1]	Duration		Reactor Type	Final SO42− [M]	Reference
A. thiooxidans DSM 504	20	35 days	4 L batch reactor		0.42	(Gourdon & Funtowicz 1998)
A. thiooxidans PTCC 1717	12.5	8 days	250 mL Erlenmeyer flasks		0.31	(Heydarian et al. 2025)
A. thiooxidans IFO 13724	10	8 days	Batch reactor with 3 L working volume		0.26	(Konishi et al. 1995)
A. thiooxidans	15	16 days	250 mL Erlenmeyer flasks		0.41	(Naseri & Mousavi 2024)
A. thiooxidans DSM 504, A. caldus DSM 9466	60	28 days	2 L batch reactor operated in semi-continuous mode		1.20	(Kremser et al. 2022)
A. thiooxidans AZ11	50	13 days	Shake flasks		0.77	(E. Y. Lee et al. 2006)
Isolated A. thiooxidans	9	unknown	Shake flasks		0.25	(Liu et al. 2004)
A. thiooxidans DSM 504	35	14 days	Stirred-tank batch reactor		1.08	Current study

Within the scope of this investigation, the importance of sulfur dosage, initial phosphate (PO₄³) concentration, and inoculum size on the microbial conversion efficiency of sulfur to SO₄²⁻ by A. thiooxidans was determined using design of experiments (DoE) methodology, with the aim of achieving high conversion rates. Understanding the effect of the investigated factors is essential for further process optimization and for identifying potential limitations in terms of acid strength and rate of reaction. Additionally, to improve process sustainability, the feasibility of using an industrial by-product from a pulp manufacturer as a sulfur source was evaluated. To facilitate future online SO₄²⁻ approximations, a correlation between measured conductivity and SO₄²⁻ concentration was explored.

Materials and methods

Design of experiments (DoE)

The DoE was conducted using the software Modde® Pro (Sartorius, Göttingen, Germany) investigating the effect of three factors, with three quantitative levels each, on three defined responses. A D-optimal quadratic model was then fitted using multiple linear regression (MLR). The factors sulfur dosage, PO₄³⁻ concentration, and inoculum size were varied according to Table 2. These levels were chosen on the basis of previous experiments conducted. Subsequently, pH, optical density (OD600), and SO₄²⁻ concentration were measured as responses. As a time point, the final sample taken on day 14 was used for statistical analysis.

Table 2.

Experimental trials

Factor	Unit	Low	Center	High
Sulfur dosage	g L−1	20	27	35
PO43−	mg L−1	25	35	45
Inoculum	(v/v) %	3	5	10

In Fig. 1 the even distribution of the 18 conducted trials is illustrated. Three trials were replicates to provide data in terms of statistical reproducibility.

Fig. 1.

3D representation of the distribution of conducted trials. Grey spheres indicate replicates and turquoise spheres indicate single runs. The numbers correspond to the order of which the trials were conducted in

Experimental set-up

To generate the inoculum, a registered culture of A. thiooxidans DSM 504 was cultivated as described by Frueholz, Habermaier et al. (2025) (Frueholz, Spiess, et al., 2025); however, without addition of yeast. The cultivation was performed in a 2 L glass reactor (Glasgerätebau Ochs, Bovenden, DE) stirred with a magnetic stirrer (IKA® RCT, Staufen, DE), with air supplied by an air pump (Schego GmbH, Offenbach, DE). After a cultivation period of 1 week, the bacterial cells were counted upon centrifugation at 778 × g for two minutes (Rotina 380, Hettich, Tuttlingen, DE) using a Neubauer Improved counting chamber (Paul Marienfeld GmbH, Lauda-Königshofen, DE), to ensure a consistent inoculum density.

For each trial run, a double-jacketed stirred-tank reactor (Schmizo AG, Oftringen, CH) with a working volume of three liters was set up as described previously (Frueholz, Habermaier, et al., 2025). The aeration was set to 90 L h⁻¹ for each run, and the stirring speed was fixed at 300 rpm. The cultivation medium for each trial was prepared as described previously (Frueholz, Spiess, et al., 2025), without the addition of yeast, and with the adjusted concentration of KH₂PO₄, to match the concentrations defined in Table 2. The pH of the cultivation medium remained unadjusted and was around 4.5. Commercially available sulfur was added in the form of technical elemental sulfur (finely ground with a purity of ≥ 99.5%), which was autoclaved separately at 105 °C for 30 min. A fixed inoculum volume with a set cell concentration of approximately 3.5 × 10⁸ cells mL⁻¹ was added at the beginning of each trial, and the reactor was left to run for 2 weeks. The cultivation duration was based on previous experiments, which were determined to be a sufficient time period to observe a levelling off of bacterial growth. The inoculum for each trial was produced in a fresh batch and was not taken from a previous trial run to avoid any effects due to bacterial adaptation. Samples were withdrawn regularly to measure SO₄²⁻, PO₄³⁻, OD600, pH, and conductivity. Sulfur-to-SO₄²⁻ conversion rate was determined following Eq. 1 and a correlation between the conductivity and SO₄²⁻ values was established. To verify no abiotic processes influence the entire experimental set-up, an abiotic trial was conducted. For this, 35 g L⁻¹ autoclaved elemental sulfur was added to sterile cultivation medium with an initial PO₄³⁻ concentration of 35 mg L⁻¹.

$$\text{Equation 1:}\frac{{\left[{\mathrm{SO}}_4^{2-}\right]}_{\text{Day 14}}-{\left[{\mathrm{SO}}_4^{2-}\right]}_{\text{Day 0}}}{\text{Sulfur dosage}\left[\frac{\mathrm{mol}}{\mathrm{L}}\right]}\times 100\%=\text{Sulfur conversion rate}$$

One final trial run was set up using biological sulfur as the sulfur source, derived from a pulp manufacturer. The biological sulfur was autoclaved according to the conditions described above. The trial was conducted with 35 g L⁻¹ biological sulfur, 35 mg L⁻¹ PO₄³⁻, and 3% inoculum. The results of this trial were not included in the statistical analysis; the trial was conducted solely to elucidate the possibility of replacing the commercial sulfur source with waste-derived sulfur.

For data analysis, OD600 was converted to cell count mL⁻¹ using a previously established relationship:

$$\text{Equation 2:}{\mathrm{OD}}_{600}\times 2.55019\times {10}^9=\text{cell count}\left[{\mathrm{mL}}^{-1}\right]$$

To further convert cell count into cell dry weight [g] the conversion factor of 2.00 × 10^–13 g/cell was applied (Fazzini et al. 2013).

Normalized reaction rate [q] was calculated using the cell dry weight from above as a value for biomass:

$$\text{Equation 3:}\frac{\mathrm{d}[{\mathrm{SO}}_4^{2-}]}{\mathrm{d}t·\text{biomass}}=q$$

Analytical procedures

The pH and conductivity of the samples were measured using the multiparameter benchtop meter inoLab® Multi 9310 IDS (Xylem Analytics, Weilheim, DE) with the respective electrodes. OD600 was measured photometrically at 600 nm (NANOCOLOR® VIS II, Macherey-Nagel GmbH, Düren, DE) upon centrifugation at 778 × g for two minutes. SO₄²⁻ concentration was determined following the procedure as described elsewhere (Kolmert & Hallberg 2000) and PO₄³⁻ concentration was measured using the photometric test kit NANOCOLOR 1–77 (Macherey-Nagel GmbH, Düren, DE).

Results

Statistical analysis

In Table 3 each trial is listed along with the measured pH, final SO₄²⁻ concentration, and OD600 of the sample taken on day 14. Additionally, the sulfur conversion rate was calculated for each run, ranging from a minimum of 61% to complete conversion. The results of the ANOVA analysis are displayed in Table S1 in the supplementary material, in which pH and SO₄²⁻ concentration were found to be significant responses. Further statistical analysis was conducted to compare data of timepoints other than day 14; however, no statistical significance was determined for any of the measured responses.

Table 3.

List of trials with defined factors and corresponding responses measured on day 14. Sulfur conversion rate is calculated for each trial

Trial No	Sulfur [g L−1]	Phosphate [mg L−1]	Inoculum v/v [%]	pH	SO42− [mol L−1]	OD600	Sulfur conversion [%]
1	27	35	3	0.47	0.71	0.297	82
2	20	25	3	0.36	0.72	0.599	108*
3	20	45	3	0.67	0.63	0.476	97
4	35	35	3	0.28	0.90	0.457	81
5	35	25	3	0.24	0.94	0.576	84
6	35	45	5	0.33	0.86	0.445	76
7	35	35	5	0.21	0.82	0.606	73
8	27	25	5	0.33	0.73	0.393	84
9	27	45	5	0.63	0.54	0.268	61
10	20	35	5	0.31	0.68	0.672	103*
11	35	25	3	0.18	1.08	0.456	97
12	27	45	10	0.36	0.60	0.381	68
13	27	35	10	0.35	0.85	0.515	97
14	20	25	10	0.38	0.70	0.528	105*
15	20	45	10	0.48	0.74	0.390	113*
16	20	45	3	0.33	0.65	0.500	100
17	35	45	5	0.24	0.85	0.362	75
18	27	35	10	0.24	0.73	0.727	82

*Conversion rates > 100% may be due to levels of elemental sulfur transferred along with the inoculum

In Fig. 2, the coefficient plot of the factors and responses is shown, indicating significant changes in the responses of pH and SO₄²⁻ concentration due to variations in sulfur supplementation. For the remaining factors investigated, no significant changes in the measured responses were detected.

Fig. 2.

Coefficient plot of the varied factors sulfur dosage (S), PO₄³⁻ concentration in the medium (P), and inoculum (Ino), and the responses pH, SO₄²⁻, and OD600 determinded on day 14. The orange-colored bars indicate significance meanwhile the green bars demonstrate insignificant effects

In Fig. 3, the main effect plot illustrates the predicted effects of varying sulfur dosage, describing a decrease in pH and an increase in SO₄²⁻ concentration with increased sulfur supplementation. Due to the bacterial oxidation of elemental sulfur to SO₄²⁻, protons are generated, which expectedly results in an acidification of the medium, corresponding to a decrease in pH.

Fig. 3.

Main effect plot describing the predicted effect of an increased sulfur dosage on the pH and SO₄²⁻ concentration of the samples taken on day 14. Dashed lines represent upper and lower confidence intervals

Course of pH, SO₄²⁻ concentration, and cell growth over time

The time-series data of the different trials categorized by the three different supplemented levels of sulfur are shown in the proceeding figures. Figure 4 displays the mean course of pH over time, demonstrating a general decrease in pH for all sulfur levels. The initial pH is lower than the pH of the pure medium, as the addition of inoculum acidifies the entire reactor solution. Standard deviations are especially high for data points on days 0 and 2, which may be due to the biological variability, especially at the beginning of each trial, as microorganisms enter their exponential growth phase.

Fig. 4.

Scatter plot depicting mean pH values over time categorized by supplemented sulfur level (n = 6). Error bars represent standard deviation

Corresponding to the decrease in pH, in Fig. 5A, the mean SO₄²⁻ concentration measured at various time points is plotted and categorized by the amount of sulfur supplemented. The concentration of SO₄²⁻ increased over time, and at the highest sulfur dosage of 35 g L⁻¹, an evident increase in SO₄²⁻ concentration from day 7 on was observed. Nevertheless, the mean sulfur-to-SO₄²⁻ conversion rate at 35 g L⁻¹ was calculated to be 81 ± 9%. For the sulfur dosages of 20 g L⁻¹, 100% sulfur conversion is achieved after 14 days. Meanwhile, for the sulfur supplementation of 27 g L⁻¹, the rate of oxidation levels off after day 11, which may be due to low sulfur availability within the solution. For the control, no increase in SO₄²⁻ could be determined over the entire course of the experiment as it remained constant at a concentration of 0.007 M. In Fig. 5B, the OD600 is plotted as mean values over time for each level of supplemented sulfur. High variability in OD600 measurements lies within the nature of biological processes, as growth rates slightly differ between trials. Nevertheless, the mean rate of cellular growth follows a similar pattern to the rate of sulfur oxidation; however, in terms of microbial growth, no distinction was observed between the different levels of supplemented sulfur.

Fig. 5.

A Scatter plot depicting the course of the mean SO₄²⁻ concentration (n = 6) measured in the samples during the course of the 14-day cultivation period, categorized by the amount of sulfur supplemented. Error bars indicate standard deviation. The control represents an abiotic experiment with a supplemented sulfur level of 35 g L⁻¹. B Mean OD600 measurements (n = 6) over time indicate cell growth during the 14-day cultivation period with error bars indicating standard deviation

To assess the rate of reaction of the conversion of elemental sulfur to SO₄²⁻, the linear region of Fig. 5A, ranging from measured SO₄²⁻ concentration from day 2 to 11, was fitted for each sulfur concentration in Fig. 6. The highest level of sulfur resulted in the highest production rate of 0.08 M SO₄²⁻ per day, compared to a lower production rate of 0.06 M when supplemented with 20 g L⁻¹ of sulfur.

Fig. 6.

Linear fits, indicated by red line, for SO₄²⁻ concentration over time using data from six individual replicates from day 2 to 11 categorized according to amount of sulfur supplemented: A 20 g L⁻¹, B 27 g L⁻¹, C 35 g L⁻¹. The dark blue region indicates the confidence band and the light blue region indicates the prediction band, both at a level of 95%

To normalize the rate of sulfur oxidation in regard to cell growth, OD600 was converted to cell count using Eq. 2 and further converted into dry cell mass. The derivative of Fig. 5A was determined and divided by the respective mean cell dry mass according to Eq. 3 at each time point with the results being plotted in Fig. 7. The cellular efficiency of the microorganisms grown in the bioreactor is fastest between day 0 and 2 and then levels off to a constant level after day 4.

Fig. 7.

Mean (n = 6) rate of SO₄²⁻ production per day normalized by cell dry mass in 3 L cell suspension over the course of time calculated for each level of sulfur

Course of PO₄³⁻ over time

The PO₄³⁻ concentration in each sample was measured along with the other responses; however, due to issues with model validity, it was not included in the statistical analysis. In Fig. 8, the PO₄³⁻ concentration determined in samples taken at different time points is shown for each initial PO₄³⁻ level. Regardless of the initial PO₄³⁻ concentration in the medium, the PO₄³⁻ concentration determined in the samples decreased over time, suggesting cellular PO₄³⁻ uptake. For PO₄³⁻ levels of 25 mg L⁻¹ and 35 mg L⁻¹, no PO₄³⁻ was detected on day 11 or day 14. For the highest level of 45 mg L⁻¹, an average of 0.96 mg L⁻¹ of PO₄³⁻ in solution was measured on day 14. Additionally, the abiotic experiment found no change in PO₄³⁻ concentration measured in the medium.

Fig. 8.

PO₄³⁻ concentration in samples over time of different initial PO₄³⁻ in medium. Colored columns represent mean values of trials with same PO₄³⁻ supplementation and error bars indicate standard deviation. Grey columns are the abiotic control using an initial PO₄³⁻ concentration of 35 mg L.⁻¹

Biological sulfur trial

One trial was conducted using biogenic sulfur to demonstrate the feasibility of using an alternative sulfur source to replace commercial elemental sulfur. The results for the measured parameters are shown in Fig. 9, indicating bacterial growth, PO₄³⁻ uptake, and a decrease in pH comparable to the trials conducted using commercial sulfur. However, only a final SO₄²⁻ concentration of 0.75 M was achieved.

Fig. 9.

A Course of SO₄²⁻ concentration and pH over time. B Course of OD600 and PO₄³⁻ concentration over time. Data is taken from one trial using 35 g L⁻¹ biological sulfur, 35 mg L⁻¹ PO₄³⁻, and an initial inoculum volume of 3%

Correlation between SO₄²⁻ concentration and conductivity

Given that measurement of the SO₄²⁻ concentration using photometric turbidity methods is time-intensive and laborious, the relationship between conductivity and the SO₄²⁻ concentration was investigated. With rising SO₄²⁻ concentration, conductivity increases, and since no other interfering ions are expected to be produced during the microbial process, a linear correlation was expected. In Fig. 10, the data for these measurements are plotted, and a line of best fit was calculated, indicating a high degree of correlation with R-Square close to 1. For upcoming investigations, online conductivity measurements can now be used to provide close approximations of the rate of microbial sulfur oxidation.

Fig. 10.

Scatter plot expressing the correlation between SO₄²⁻ concentration determined in the samples and the corresponding conductivity measurements

Discussion

Previous studies investigated the importance of parameters in sulfur oxidation and reported comparable findings. Liu et al. used response surface methodology to evaluate the effects of different nutrients, along with elemental sulfur, on the biological oxidation of sulfur to SO₄²⁻ by A. thiooxidans and found the level of sulfur to correlate with the SO₄²⁻ concentration (Liu et al. 2004). Heydarian et al. investigated the effects of sulfur dosage, inoculum size, and initial pH on the final SO₄²⁻ concentration and pH, revealing a correlation between sulfur oxidation and sulfur dosage (Heydarian et al. 2025). Although the significance of sulfur supplementation is not a novel finding, previous studies focused on relatively low sulfur concentrations. The present results demonstrate that the same effects persist at higher concentrations. Additionally, the scale at which these experiments were conducted is not trivial, as extrapolation from shake-flask systems is often unreliable. In this study, all experiments were carried out in a 5-L reactor, demonstrating the feasibility of oxidizing elevated sulfur concentrations within 14 days.

In terms of the amount of sulfur dosage, at high levels, a major limitation is the intrinsic hydrophobicity of the commercial sulfur feedstock itself. At the highest level of 35 g L⁻¹ of sulfur, clumps of sulfur tended to accumulate on the reactor vessel walls, reducing its bioavailability and thus preventing complete conversion. Therefore, no higher sulfur concentration was tried out within this investigation. Further investigations are planned to optimize the mode of sulfur addition in order to maximize sulfur bioavailability in suspension. Additionally, even though the bacterial growth rate decreases after the first week of cultivation, the rate of sulfur oxidation normalized by biomass remains constant, indicating the need for further reactor experiments with a longer cultivation duration in order to further maximize SO₄²⁻ concentration and sulfur-to-SO₄²⁻ conversion. Regarding the rate of sulfur oxidation, Gourdon and Funtowicz found a linear relationship between SO₄²⁻ concentration and time from day 5 to 20, with the rate decreasing close to zero after day 20. The authors postulated the inhibiting effect of a high SO₄²⁻ concentration; however, their final SO₄²⁻ concentration was around 0.42 M, thus being far below the maximum concentration of the present investigation (Gourdon & Funtowicz 1998). In Fig. 6 no such inhibition is found at a SO₄²⁻ concentration of 0.42 M; however, as displayed in Fig. 5A, a levelling off of the oxidation rate seems to occur after day 12, especially for the lower sulfur levels, as approaching a sulfur conversion rate of 100%. Here too, a longer cultivation period using a sulfur level of 35 g L⁻¹ may give insight into the form and extent of inhibition occurring.

Regarding the effect of the initially supplemented PO₄³⁻, no significant effect was found on the sulfur oxidation rate or on the microbial growth rate, suggesting that the investigated concentrations were sufficient for microbial growth. An excessive surplus of PO₄³⁻ has been shown to have an inhibitory effect on sulfur oxidation (Vogler & Umbreit 1942); by contrast, PO₄³⁻ starvation conditions in the closely related species A. ferrooxidans have been found to inhibit cell growth and reduce the rate of iron oxidation (Seeger & Jerez 1993).

The experiment using biogenic sulfur instead of commercial sulfur solely yielded a sulfur conversion of 64%, which highlights the need for further trials and optimization. Previously, the use of biological sulfur or dispersed sulfur has yielded improved results compared with commercial sulfur due to its improved solubility (Inaba et al. 2021; Seidel et al. 2006). To improve the sulfur conversion rate, an adaptation period may be important to allow microorganisms to adapt to the new sulfur source. Furthermore, the particle size may play a role, as smaller particles solubilize faster and the elemental sulfur may be ground finer compared to the biological sulfur. Lastly, the composition of the biological sulfur may matter since impurities might influence conversion or simply inflate the measured sulfur content.

The biologically produced H₂SO₄ can be applied in a variety of industrial applications including waste-management scenarios, replacing commercially available diluted H₂SO₄. For instance, the extraction of critical metals from spent lithium-ion batteries has been previously described (Baniasadi et al. 2025; Heydarian et al. 2025; Naseri & Mousavi 2024); however, the reported SO₄²⁻ concentrations in the biogenic acid ranged from 30 to 40 g L⁻¹, which is approximately half of the concentration of the biogenic acid produced in the present investigation. Higher acid strength allows the use of higher pulp densities, directly affecting the economic feasibility of such processes. Selective leaching of zinc from electric-arc-furnace dust (Frueholz, Spiess, et al., 2025) or the extraction of phosphorus from sewage sludge incineration ash (Fang et al. 2021) are further examples where the biogenic acid could be used as lixiviant to recover critical resources from waste materials.

Conclusion

Within the scope of this investigation, a significant effect of the sulfur dosage on the SO₄²⁻ concentration after a 14-day cultivation period, measured in the biological lixiviant, was demonstrated using DoE methodology. The mean sulfur-to-SO₄²⁻ conversion rate at a sulfur dosage of 35 g L⁻¹ was 81 ± 9% and the highest final SO₄²⁻ molarity achieved during the entire experiment was 1.08 M with a corresponding pH of 0.18. The other factors varied, initial PO₄³⁻ and inoculum volume, did not have a significant impact on the measured SO₄²⁻, OD600, or pH. A general depletion of PO₄³⁻ concentration in the lixiviant was observed for each trial, and OD600 increased steadily over the course of each trial. The feasibility of using an industrial by-product as an alternative sulfur source was demonstrated. Lastly, a correlation between the measured conductivity and the SO₄²⁻ concentration was established.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

R.F. wrote the main manuscript text, analyzed data and prepared all figures. R.F., C.H., and L.H. carried out experimental work and measurements. S.S. and M.H. supervised the project and provided advice. C.H. provided advice regarding data analysis. G.G. provided critical feedback. All authors reviewed the manuscript and provided feedback.

Funding

The project is co-funded by the European Regional Development Fund (programme Interreg Austria–Czechia 2021–2027), project ATCZ00043, PHOS4PLANT (From Waste to Resource—Recycling Sewage Sludge Ash into Phosphate-rich Plant Fertilizer). Furthermore, the authors gratefully acknowledge the funding support of K1-MET GmbH, metallurgical competence center. K1-MET is a COMET Centre within the COMET—Competence Centers for Excellent Technologies Programme and funded by the Federal Ministry of Innovation, Mobility and Infrastructure Republic of Austria, the Federal Ministry of Economy, Energy and Tourism Republic of Austria, the provinces Upper Austria, Styria, and Tyrol as well as the the Steirische Wirtschaftsförderungs­gesellschaft m.b.H. (SFG) and the Standortagentur Tyrol. Furthermore, we thank Upper Austrian Research GmbH for the continuous support. The COMET Programme is managed by FFG.

Data availability

All data generated or analyzed during this investigation are included within the article or found in the supplementary material.

Declarations

Ethical approval

Not applicable.

Competing interests

The authors declare no competing interests.