An Integrated Sulfide Removal Approach from a Substrate for Biogas Production and the Simultaneous Production of Electricity

Abstract

Sulfides are frequently encountered in natural mineral water and different wastewater streams, and their presence significantly impedes subsequent water treatment or utilization. Sulfide removal, or at least its reduction, can be accomplished in different ways, but there is one straightforward method where sulfide is captured on a carbon-based sorbent, with the consequent sorbent regeneration producing electricity in liquid fuel cell mode. This multi-functional approach combines sulfide removal, energy generation, and water pre-treatment for various applications. The present work aims to show sulfide removal from sulfide-containing wastewater streams from alcohol and beverage manufacturing. The clean water could be used for biogas production. Sorbent regeneration was performed in fuel cell mode and was accompanied by electricity production. The experiments, conducted in a liquid-phase fuel cell, used electrode compartments that were separated by an anion-exchange membrane. Electroconductive charcoal, produced via the pyrolysis of residues from tire production and doped with zinc oxide, was used as a sorbent. The experimental treatments of vinasse, whey, and stillage for sulfide removal by this method show the sustainable performance of the sorbent for up to twelve consecutive runs. The biogas yield produced from vinasse was increased more than three times for the treated substrate compared to the reference case.

1. Introduction

Hydrogen sulfide is found in many aspects of nature, including as a component of mineral waters, a by-product of organic matter biodegradation, and a waste product of various industrial applications, like oil processing, mining, tanning, and pulp and paper processing [1]. The presence of hydrogen sulfide in natural water and waste streams in industry is undesirable because of its high toxicity and corrosivity for the metal elements of industrial equipment. A significant source of hydrogen sulfide exists in the Black Sea waters at a certain depth [2]. Hydrogen sulfide is harmful to the environment even at very low concentrations because it is extremely toxic, and its presence in waste streams used for biogas production by anaerobic digestion is undesirable because of its harmful effects on the involved microbial consortia [3,4].

The low concentrations of hydrogen sulfide and the very large amounts of polluted air and water to be treated make its removal very costly. Different types of removal include chemical (oxidation, adsorption, absorption, and sedimentation) [5,6,7,8], physical (membrane separation, condensation, and ion exchange) [9,10], and biological methods (biofiltration, where hydrogen sulfide is oxidized to sulfate by thiobacteria) [11,12].

Among the chemical methods, oxidation and absorption are expensive because of the huge industrial streams that need to be processed and the associated regeneration of the sorbents [13]. An additional obstacle to this approach is the necessity of removal and treatment of the heavy metal sulfide residues, as they have adverse effects on the environment and human health.

There are various methods for mitigating the harmful effects of hydrogen sulfide on anaerobic digestion for biogas production. The majority of them are focused on controlling anaerobic processes, but not as the preliminary removal method [4,14]. A comprehensive review has discussed methods for capturing hydrogen sulfide by using ionic liquids, eutectic solvents, zeolites, etc. [7], which can be used for preliminary sulfide removal from waste streams, although there is a lack of data in the literature for such operations prior to their use for biogas production.

The present study proposes another simple and energy-saving mechanism to treat waste streams with low concentrations of hydrogen sulfide in a simple regeneration process, which consists of capturing sulfide by chemosorption and subsequent sorbent regeneration in fuel cell mode using the sorbed sulfide as a reductor.

Similar attempts were made for waste gas streams—where hydrogen sulfide was oxidized to elemental sulfur [15]—and for aqueous solutions in a liquid phase fuel cell [16].

Recently, an integrated method for hydrogen sulfide adsorption on charcoal doped with zinc oxide was proposed [17]. The regeneration of the adsorbent was accomplished in fuel cell mode in an aqueous medium. The electroconductive charcoal served as the anode, and the adsorbed pollutant, zinc sulfide, was directly oxidized and washed out as sulfate or sulfite. Thus, the sorbent capacity was restored with the generation of electricity. The principal sketch of the proposed process is shown in Figure 1.

The advantages of this approach are that the processes are straightforward and involve a few simple operations. Energy saving also occurs because generating electromotive force requires low energy consumption.

The present study aims to remove sulfide from sulfide-containing waste streams via chemosorption with subsequent sorbent regeneration in a fuel cell mode and analyze its effect on biogas generation during anaerobic digestion. This work shows experimental results for sulfide removal from three substrates of natural origin via chemosorption on zinc oxide, which was integrated with sorbent regeneration prior to biogas production. The purified substrate is tested for biogas production, and comparisons of biogas yield with and without preliminary sulfide removal are given.

2. Materials and Methods

2.1. Fuel Cell Experiments

Experiments on sorption and fuel cell regeneration of the sorbent, conducted in batch and continuous processes in a rectangular fuel cell, were carried out in 2024 and 2025, as shown in Figure 2. The anodic space was packed with sorbent particles of charcoal doped with zinc oxide. The sorbent was prepared via the pyrolysis of residues from tire production with the addition of an aqueous solution of zinc acetate. On average, the size of the sorbent particles was 0.1 cm. The prepared chemosorbent is electroconductive and therefore contributes to an increased anode area. The electrodes were made out of 10 × 10 cm sintered graphite square plates (a Cometech OOD production).

Both electrode spaces in the fuel cell had a volume of 50 m each.

The fuel cell compartments were separated by an anion-exchange membrane (Selemion, Asahi Glass Co., Tokyo, Japan). As the first step, the sorbent was continuously saturated in the anode space by sulfide contained in the substrate solutions (which were introduced into the anodic space by a peristaltic pump with a flow rate of 0.02 to 0.4 dm³/h) in an open circuit. Saturation ended when the open circuit voltage of the fuel cell reached a constant value for a certain time, e.g., three successive measurements in 15 min. Samples from the inlet and outlet solutions for experiments of sorbent saturation were taken and analyzed for sulfide. The sorption capacity S (wt.%) for a sorbent with mass G was calculated via integration over time for the saturation of the differences between the inlet and outlet sulfide concentrations for the used flow rate Q.

$$S-{\displaystyle \frac{1}{G}}Q{\int }_0^t(c_{in}-c_{out}).dt$$

(1)

The sorption capacity was estimated as 0.14 % ± 0.04 (wt.).

For the second step, the saturated sorbent was regenerated via a closed-circuit process where anode oxidation of the retained sulfide occurred in a fuel cell mode. This process was carried out in two modes: static (i.e., batch process) and continuous wash-out with a solution of the supporting electrolyte. Continuous fuel cell discharge occurred during wash-out of the fuel cell’s anode space via the supporting electrolyte solution at a flow rate of 0.2 to 0.4 dm³/h. Oxygen was introduced in the cathode space either in continuous or batch experiments by continuously passing aerated solutions of the supporting electrolyte through it at a flow rate of 0.35 dm³/h. A sodium chloride solution (16 g dm⁻³) was used as the supporting electrolyte and for wash-out of the sorbent in the continuous fuel cell experiments.

The regeneration of the saturated sorbent was accompanied by the generation of electromotive force. Anode reactions involving hydroxylic anion exchange with their standard potentials are considered. The expected anode processes in the fuel cell are as follows:

First step:

ZnO + 2H⁺ + S²⁻ = ZnS + H₂O-chemo-sorption

Second step:

ZnS + 8OH⁻ − 8e = ZnO + SO₄²⁻ + 3H₂O + 2H⁺, electrochemical oxidation, E₀ = −0.69 V,

However, different sulfide oxidation reactions are possible, as they depend on the initial sulfide concentration in the solute and the possibility of bulk oxidation, competing with the anode processes. Some of these reactions are shown in Table 1.

The electrochemical cathode reaction is as follows:

O₂ + H₂O + 4e⁻ = 4OH⁻, E₀ = 0.401 V.

There are dual benefits to this process: sulfide removal and energy production.

The treated substrates used for biogas production included residual stillage from ethanol distillation, residual vinasse from wine production, and lactate-containing whey; all were taken from industrial processes in Bulgarian enterprises.

Their contents were determined by HPLC and were reported as follows [18]:

Stillage, g dm⁻³: glucose, 0.14; xylose, 0.07; mannose, 0.27; galactose, 0.02; arabinose, 0.04; cellobiose, 0.26; lactic acid, 7.68; acetic acid, 0.02; propionic acid, 1.56; ethanol, 0.66.

Vinasse, g dm⁻³: glucose, 0.3; xylose, 1.12; mannose, 0.2; cellobiose, 0.26; lactic acid, 1.2; ethanol, 2.82.

Whey, g dm⁻³: lactic acid, 3.01.

The organic substrates were purged with nitrogen prior to feeding the sorbent.

The pH values of the initial solutions ranged from 7.3 to 12.6, depending on the sulfide concentration, which varied from 18 to 400 mg dm⁻³.

Twelve experiments were carried out. Polarization curves were created after each saturation trial at different current densities, varying the external resistance of the circuit. Then, experiments on sorbent regeneration via cell discharge through selected ohmic resistances in continuous or batch processes were carried out. The electric current was calculated using Ohm’s law from the measured cell tension and the external resistance, and the voltage measurements were obtained with an accuracy of ±1 mV.

The amounts of electrochemically oxidized sulfide were computed from the measured electric current using Faraday’s law, shown in Equation (2), and compared to the sulfide depletion determined through the analyses.

$${\displaystyle \frac{\mathrm{m}}{\mathrm{t}}}={\displaystyle \frac{\mathrm{M}\mathrm{i}}{\mathrm{n}\mathrm{F}}},\mathrm{m}={\displaystyle \frac{\mathrm{M}}{\mathrm{n}\mathrm{F}}}{\int }_0^{\mathrm{t}}\mathrm{i}.\mathrm{d}\mathrm{t}$$

(2)

Here,

• i—electric current, A;
• m—mass of reacting substance, g;
• t—time, s;
• M—molar mass of reacting substance, g mol⁻¹;
• n—number of exchanged electrons;
• F = 96,484 C mol⁻¹, Faraday constant.

Sulfide can participate in various redox reactions with various products with different numbers of exchanged electrons, becoming more diverse at higher sulfide concentrations. A short list of such reactions is shown in Table 1.

Table 1. Short excerpt of redox reactions involving sulfide oxidation [19].

No.	Reversible Anode Reaction	Number of Exchanged Electrons, n	Standard Electrode Potential, V, 25 °C
1	SO32− + 3H2O + 6e = S2− + 6OH−	6	−0.66
2	SO42− + H2O + 2e = SO32− + 2OH−	2	−0.91
3	S22− + 2e = 2S2−	1	−0.524
4	S + 2e = S2−	2	−0.480
5	S2O32− + 6H+ + 8e = 2S2− + 3H2O	4	−0.006
6	SO42− + 4H2O + 8e = S2− + 8OH−	8	−0,693

Table 1. Short excerpt of redox reactions involving sulfide oxidation [19].

No.	Reversible Anode Reaction	Number of Exchanged Electrons, n	Standard Electrode Potential, V, 25 °C
1	SO32− + 3H2O + 6e = S2− + 6OH−	6	−0.66
2	SO42− + H2O + 2e = SO32− + 2OH−	2	−0.91
3	S22− + 2e = 2S2−	1	−0.524
4	S + 2e = S2−	2	−0.480
5	S2O32− + 6H+ + 8e = 2S2− + 3H2O	4	−0.006
6	SO42− + 4H2O + 8e = S2− + 8OH−	8	−0,693

2.2. Analyses

Quantitative analyses for sulfide in the substrates before and after sorption and during regeneration were carried out photometrically with N,N-dimethylnphenylenediamine in the presence of Fe (III) to form methylene blue at a 667 nm wavelength [20]. For this purpose, the UV-1600PC UV/Vis spectrophotometer was used. In addition, a straight calibration line for the 0–0.67 mg dm⁻³ concentration range was used, and the samples were properly diluted to fit the linear range of the calibration line. The presence of sulfate and sulfite was proven qualitatively by the addition of barium chloride. Barium sulfite is soluble after the addition of mineral acid, whereas barium sulfate is not. The presence of thiosulfate was tested by the addition of ferric salts to give a purple color. Polysulfides give clear solutions (yellow for S₂²⁻ and green for S₃²⁻) with the deposition of colloidal sulfur after acid addition. The chemicals used were of p.a. grade.

2.3. XRD Measurements

At the Geological Institute of the Bulgarian Academy of Sciences, powder X-ray diffraction measurements of the sorbent prior to and after 12 experiments were carried out using HUBER Image Plate Guinier Camera G670 in an asymmetric transmission mode with a Ge monochromator on the primary beam, providing pure Cu Kα₁ radiation (λ = 1.540598 Å). X-ray diffraction data were simultaneously taken in the range of 4 to 100° two theta with a step size of 0.005° 2θ. The diffraction data were processed with the Match! software package for phase identification by CRYSTAL IMPACT, Bonn, Germany [21]. Phase identification and semi-quantitative phase analysis were performed using the reference patterns in the Powder Diffraction File database of the International Center for Diffraction Data (ICDD PDF-2) [22] and the Crystallography Open Database (COD) [23,24].

2.4. Biogas Production

The experiments for biogas production from the treated organic substrates, carried out in 2025, were accomplished in a lab-scale mode, where 150 mL of the treated substrates was mixed with 300 mL of activated sludge from the wastewater treatment plant. The sealed flasks containing the samples were kept at 32 °C in a water bath under static conditions, and the released biogas was collected in a gas holder. The experiments lasted for 30–35 days. Parallel reference tests with substrates without treatment were also carried out. A sketch of the lab-scale equipment for biogas production is shown in Figure 3.

3. Results and Discussion

3.1. The Sorbent Composition

The XRD diagrams of the sorbent used before and after 12 runs are presented in Figure 4, which includes Miller indices for the detected crystal compounds. A quantitative analysis of the XRD diagrams shows the following composition of the sorbent before use (wt.%): sphalerite (ZnS, 52.4); wurtzite (ZnS, 24.4); calcite (CaCO₃, 10.1); quartz (SiO₂, 6.8); and zincite (ZnO, 6.3). After 12 runs, the amount of wurtzite dropped considerably to 16.6% and zincite was slightly reduced to 5.1%. This observation can be explained by the wash-out of sulfite and sulfate formed after the oxidation of both wurtzite and zincite in the sorbent during the fuel cell operations. Sphalerite was virtually not released. The sorption capacity remained almost constant (at about 0.14 %wt.) because its active component, zincite, was slightly reduced. Halite (i.e., NaCl) was detected in the sorbent used, which can be explained by contamination from the supporting electrolyte and the substrates.

3.2. Sulfide Sorption and Removal

The experimental conditions and results for sulfide sorption and removal from the liquid phases at different substrates are shown in

Table 2. The initial (or inlet) and residual sulfide concentrations limit the open circuit voltage and maximum power densities determined for different organic substrates after saturation.

Substrate	Initial Sulfide Concentration, mg dm−3	Residual Sulfide Concentration, mg dm−3	Initial Open Circuit Voltage, V	Maximum Power Density, W m−2	Process
Vinasse	54.3	<1.0	0.330	0.132	Continuous feed
Vinasse	62.3	<1.0	0.373	0.159	Batch
Whey	358	0.9	0.473	0.194	Batch
Whey	392	<0.1	0.513	0.219	Batch
Stillage	18	<0.1	0.300	0.169	Batch
Stillage	39.7	<0.5	0.342	0.169	Batch
Stillage	79.1	<0.5	0.454	0.384	Batch
Average			0.400 ± 0.080	0.204 ± 0.080

. There is no clear correlation between the substrate origin, the initial sulfide concentration, and the open circuit voltage. However, it was found that the sulfide anions are almost completely captured by the sorbent (Table 2). The maximum power densities during regeneration are stable, regardless of the substrate origin and sulfide feeding concentration.

A polarization line for a fuel cell operating with an already saturated sorbent is shown in Figure 5. The straight line indicates no overpotential at low currents, nor mass transfer limitations at higher currents.

A discharging curve of the sorbent in a fuel cell mode for the sorbent regeneration step under static conditions is shown in Figure 6. The electric current reaches almost-zero values, i.e., after sorbent regeneration, there is virtually no more sulfide retained on the sorbent, which was confirmed through analyses. The analytically determined amounts of sulfide removed after the runs under static conditions correspond to the initially introduced amounts.

The measured and calculated sulfide depletion values during regeneration in fuel cell mode can be affected by the competitive oxidation of wurtzite on the anode. A comparison of the measured current densities and the sorption capacities during the first 11 runs is shown in Figure 7, where after a drop in current density in the first two runs, the value remains constant, i.e., 4.6 ± 1.0 A/m². This enables us to confirm that the main effect on the electric current is due to the zinc sulfide formed on zincite’s active centers, whereas the released wurtzite was dissolved and removed during the first two experiments. An additional indication for this explanation is the constant sorption capacity throughout the experiments, i.e., 0.14 ± 0.05% (wt.).

The mass values of the sulfide removed by electrochemical oxidation in fuel cell mode were calculated using Faraday’s law (Equation (2)). These values were compared to the captured amounts of sulfide determined by chemical analyses. However, the calculated values for sulfide oxidation depend on the number of electrons exchanged on the anode (Table 1). Some of these results are shown in Figure 8.

It is evident from the cases of vinasse and stillage as substrates that reactions with the exchange of one electron are not realistic. The current yields calculated for reactions with one exchanged electron gave much higher amounts of depleted sulfide than those determined analytically. The oxidation of sulfite to sulfate is more probable, as shown by reaction 2 in Table 1, which involves the exchange of two electrons on the anode. This was confirmed by the sulfite and sulfate detected in the final solutions after fuel cell discharge in batch experiments or the final solutions when continuous wash-out took place. No elemental sulfur, nor polysulfide or thiosulfate, was detected in the effluents.

In the case of high initial sulfide concentrations, i.e., when whey was tested, the calculated yields of oxidized sulfide were much lower than those determined analytically.

3.3. Results for Biogas Production

A comparison of the results of biogas production before and after sulfide removal when vinasse was used as the substrate is shown in Figure 9. The accumulative biogas yield is more than three times higher than the reference case, demonstrating a net positive effect of sulfide removal.

This sulfide removal process is more advantageous compared to those for hydrogen sulfide removal from the already produced biogas [9,10,25,26], which include scrubbing by water, iron hydroxide, sodium hydroxide, and solid phase adsorption. All of these include treatment of the resulting sulfide-containing waste, which is costly and environmentally inappropriate [25]. The combined solid phase adsorption with photocatalytic oxidation is a more suitable method [26]. In our case, hydrogen sulfide is removed from the substrate, thus facilitating methane production and further biogas utilization.

4. Conclusions

A new and simple method for the removal of hydrogen sulfide and sulfide anions from natural and industrial waters is proposed. This method is based on integrated processes of adsorption (or chemosorption) and the subsequent removal of sulfide from the sorbent in fuel cell mode via electrochemical oxidation to harmless sulfite and sulfate in aqueous solutions. Sulfide removal and the subsequent utilization as energy have multiple advantages, and these methods can be combined with water pre-treatment for various purposes. Significant energy is generated in the fuel cell with this method, enabling the equipment to be self-sufficient (at least partially).

The method involves the chemosorption of sulfide to zinc oxide embedded in charcoal as a carrier and was tested with three different substrates: vinasse, stillage, and whey. Complete sulfide removal from the tested substrates was attained. The subsequent step, the oxidation of the formed zinc sulfide to sulfite and sulfate, removes these products from the sorbent and restores the sorption capacity. Hence, the purified streams can be used for different applications as they are non-toxic and non-corrosive. The positive effect of the proposed procedure was demonstrated on treated vinasse as a substrate for biogas production. The biogas yield produced from vinasse after sulfide removal was considerably increased compared to the reference case.

The proposed method has several advantages over the known processes of scrubbing, adsorption, and sedimentation due to its simplicity and the avoidance of auxiliary processes of secondary waste treatment and excessive energy spending. It does not involve in situ control for mitigation during anaerobic digestion for biogas production [3,4], nor subsequent hydrogen sulfide removal from the produced biogas [9,10,25,26].

The proposed process can be extended for the treatment of sulfide-containing mineral waters and waste streams, enabling the use of the purified water for industrial purposes.