Enhanced sulfide removal by gas stripping in a novel reactor for anaerobic wastewater treatment

Much attention has been paid to anaerobic wastewater treatment processes because they enable the recovery of energy from waste organics via methane gas. To enable anaerobic treatment, inhibition caused by ammonia, sulfide, light metals, heavy metals, and organic materials needs to be addressed. These inhibitors are often found in wastewater or produced during anaerobic digestion; therefore, the incorporation of methods to eliminate inhibitors can result in higher treatment efficiency and process stability (Chen et al. 2008; Yuan & Zhu 2016).

Sulfide in the liquid phase is toxic to methanogens and inhibits anaerobic processes (Karhadkar et al. 1990; Chen et al. 2008). Sulfide can be a component of wastewater or produced by sulfate-reducing bacteria. Sulfide and sulfate are often found in industrial wastewaters produced by the tannery, pulp, and paper industries (Wiemann et al. 1998; Hao et al. 2014). The 50% inhibition level for methanogenic activity has been reported to be 90–250 mg S/L under different pH conditions (Koster et al. 1986). Other studies have reported even higher inhibition levels. For instance, 50% inhibition of methanogenic activity was also observed at a sulfide concentration of 500 mg S/L and pH of 7.0–7.2 (Karhadkar et al. 1990) and at 650 mg S/L and pH 7.5 for acetate-utilizing methanogens (Onodera et al. 2011). Notably, free hydrogen sulfide (H₂S) is more toxic to methanogens than hydrosulfide ion (HS⁻) (Koster et al. 1986).

A strategy for minimization of free H₂S is important to successful anaerobic treatment of wastewater (Omil et al. 1995). To reduce free H₂S, pH control or additional processes for sulfide removal are required. Sulfide removal processes include chemical reactions (oxidation and precipitation), physico-chemical techniques (gas stripping), biological conversions (elemental sulfur production), and combined techniques (Chen et al. 2008; Li et al. 2015). In two-stage processes, sulfide removal can be applied via acidification in the first stage and methane production in the second stage (Stefanie et al. 1994). Acidification lowers the pH and increases the H₂S/HS⁻ ratio, allowing the gas-stripping process to reduce sulfide more effectively in the liquid phase (Omil et al. 1995; Wei et al. 2007). Although gas-stripping processes are effective, they require mechanical devices and energy consumption (Hao et al. 2014).

In this study, we investigated the use of biogas produced during anaerobic digestion for gas stripping of H₂S. To accomplish this, we developed a novel concept in an anaerobic reactor consisting of a gas-stripping stage, methane fermentation stage, and a phase separator (Onodera et al. 2022). In this system, the gas stripping and methane fermentation stages were arranged vertically and connected by a phase separator that consisted of a newly designed gas–liquid partitioning (GLP) valve. The critical function of the GLP valve is to block liquid flows between the stages while allowing biogas generated by methane fermentation in the lower segment to enter the upper segment for gas stripping. Unlike conventional gas-stripping processes (Stefanie et al. 1994), the GLP valve allows gas stripping without the need for mechanical devices (e.g., blowers and diffuser pipes) and electric power. The developed system was named an internal phase-separated reactor (IPSR). The first study of an IPSR fed with high-sulfate wastewater revealed higher H₂S removal via biogas and lower sulfide concentrations in the effluent under a high organic loading rate (OLR) of 10 kg COD/(m³ day) (Onodera et al. 2022). Thus, the IPSR may be a powerful treatment technology that can be applied to many types of wastewaters. However, little is known about the process performance with respect to various types of wastewater, such as sulfide wastewater. It is also important to continue to demonstrate the technical superiority of IPSR under different conditions.

Here, we applied the IPSR for high-sulfide wastewater and evaluated its feasibility for H₂S removal using biogas produced in the IPSR. Both reactors were operated in parallel and fed with synthetic wastewater containing sulfide at 400–600 mg S/L. The IPSR was operated without recirculation and directly fed with sulfide at 400–600 mg S/L so that the effects of the GLP valve on H₂S removal could be evaluated directly.

In the first stage, which did not contain sludge, sulfide was removed by gas stripping using biogas. The second stage, which was used for methane fermentation, was filled with mesophilic granular sludge obtained from an upflow anaerobic sludge blanket (UASB) reactor fed with synthetic wastewater. The UASB was operated at an OLR of 20 kg COD/(m³ day) and the sludge was taken on day 677. Details regarding the operating conditions and reactor performance of the UASB are presented elsewhere (Onodera et al. 2022). Operation started within 30 min of the sludge being transferred to the reactors. The total solid (TS) concentration of the sludge was 70.9 g/L, while the volatile solid (VS) concentration was 63.5 g/L (VS/TS = 0.90) and the sludge volume index was 2 mL/g TS. The total volume of the sludge was 4.0 L (284 g TS and 254 g VS for each reactor). After the reactor experiment, the sludge in the IPSR and the VS contained 163 g VS (189 g TS) and 149 g VS (199 g TS), respectively.

The reactor operating conditions are shown in Table 1. During the first period, the reactors were operated without feeding high concentrations of sulfide (Na₂S). The comparative experiment was conducted after the second period. The configurations of the IPSR and the control reactor were the same except for the mode of the phase separator (Figure 1). In the IPSR, biogas produced in the second stage moved to the first stage, after which it was released. In the control reactor, biogas was released from the middle portion of the reactor via the GLP valve (Figure 1).

The reactor was fed with synthetic wastewater under the operating conditions described in Table 1. The wastewater concentrations shown in the table were determined after mixing with sulfide solution (10×) consisting of Na₂S·9H₂O. The organic substances in the synthetic wastewater consisted of 2,400 mg/L sucrose, 1,800 mg/L CH₃COONa, 900 mL/L propanoic acid, and 600 mg/L peptone. Additionally, the wastewater contained 320 mg/L Na₂SO₄, 4,000 mg/L NaHCO₃, 150 mg/L KH₂PO₄, 150 mg/L CaCl₂·2H₂O, 400 mg/L MgCl₂·6H₂O, 300 mg/L KCl, 110 mg/L NH₄Cl, and a trace element solution. The trace elements solution consisted of 3.93 mg/L FeCl₂·4H₂O, 0.17 mg/L CoCl₂·6H₂O, 0.07 mg/L ZnCl₂, 0.06 mg/L H₃BO₃, 0.5 mg/L MnCl₂·4H₂O, 0.04 mg/L NiCl₂·6H₂O, 0.03 mg/L CuCl₂·2H₂O, and 0.03 mg/L NaMoO₄·2H₂O. This wastewater composition was based on an established method (Sekiguchi et al. 1998). The pH in the influent was adjusted to approximately 6–7 using H₂SO₄ solution, after which an anti-foaming reagent (Antifoam SI, FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) was added. The synthetic wastewater was prepared once a week and kept at room temperature (approximately 26–27 °C).

Water samples were taken from the influent, first segment effluent, and second segment effluent. The pH and oxidation–reduction potential (ORP) were determined immediately after sampling using a pH meter (9625-10D, D-74, LAQUA, Kyoto, Japan) and an ORP meter (9300-10D, D-74, LAQUA), respectively. For dichromate COD (COD_Cr) analysis, the sample was amended with H₂SO₄, then purged with N₂ to remove the sulfide. After pre-treatment of samples, the COD was measured using TNT822 and TNT823 test kits (Hach, Loveland, CO, USA). The SO₄ concentration was determined using SulfaVer4 (2106769, Hach). The COD_Cr and SO₄ were measured using a Hach water quality analyzer (DR1900, Hach). The produced biogas was collected in a gas bag after H₂S trap and the volume was measured with a wet-test gas meter (WS − 1A, Shinagawa, Tokyo, Japan). The biogas composition was determined using a gas analyzer (Biogas 5000, Geotech, Geotechnical Instruments Ltd, Leamington Spa, UK). The H₂S in the gas before H₂S trap was measured using a gas detector tube system (4HP and 4HH, GV-100S, GASTEC, Kanagawa, Japan). The biogas production was determined after H₂S trap. The volume of biogas and methane produced was calculated using standard conditions. Other analyses were conducted on the basis of standard methods (APHA 2005).

In the second period, the influent sulfide concentration was 400 mg S/L. The results showed that the performance of the IPSR was better than that of the control reactor. The COD of the second effluent was approximately 500 mg/L for the IPSR and over 1,000 mg/L for the control reactor. The biogas and CH₄ production were also lower in the control reactor than the IPSR. Moreover, the control reactor showed an increase in COD in the second effluent and considerably decreased biogas and CH₄ production in the third period. Thus, COD removal was related to biogas and CH₄ production. The pH in the second effluent was also lower, while the ORP was increased in the third period. These findings indicate that the process had collapsed in the control reactor because of sulfide inhibition. The IPSR also had a slightly higher COD concentration and lower biogas and CH₄ production when the performance in the second period was compared, but maintained superior performance compared with the control reactor. The pH was neutral and the ORP was less than −400 mV in the second effluent of the IPSR, indicating that the IPSR could tolerate high influent sulfide levels in the second (400 mg S/L) and third period (500 mg S/L).

The IPSR removed hydrogen sulfide from the wastewater by gas stripping, thereby reducing the low concentration of sulfide in the wastewater. As a result, sulfide inhibition in the methanogenic phase is expected to be alleviated. Instead of reduced sulfide concentrations in the wastewater, the biogas produced contained high concentrations of hydrogen sulfide. To enable biogas utilization for electricity generation or upgrading to biomethane, H₂S must be removed from the biogas (Choudhury et al. 2019). Various treatment methods for desulfurization of biogas have been proposed in previous studies (Lupitskyy et al. 2018). However, an appropriate technology for H₂S removal from biogas produced in an IPSR needs to be selected as a post-treatment of biogas during its application.

Several methods for removal of sulfide from wastewater have been proposed, including physical methods, chemical reactions, and biological transformations (Chen et al. 2008; Robles et al. 2020). However, many of these techniques incur additional costs because they require chemicals (e.g., precipitation) or mechanical operations (e.g., gas stripping using air or N₂). Therefore, the primary advantage of IPSR is that it is a simple and low-cost process. This is because the biogas produced by the IPSR can be used for gas stripping and introduced naturally into the reactor. Although sulfide removal in the IPSR was sufficient in this study, further sulfide removal may be difficult to control because it depends primarily on the amount of biogas produced. To enable application to wastewater with high sulfide content and high S/COD ratios, further research and development is needed to evaluate the performance of the IPSR while treating super-high sulfur wastewater. Such studies should determine the maximum feasible level that can be treated and improve gas stripping efficiency.

This study evaluated the performance of sulfide and COD removal by an IPSR. The results showed that the IPSR has unique mechanisms for H₂S removal before methane production occurs by using produced biogas. As a result, the IPSR had higher organic removal than the control reactor. Additionally, the IPSR removed twice the H₂S of the control at a sulfide concentration of 400 mg S/L. Because of its high sulfide removal, the IPSR tolerated sulfide concentrations in the influent wastewater at high levels that caused the control reactor process to collapse.

We thank Mrs Naoko Okubo for support with the experiments. We thank Edanz (https://jp.edanz.com/ac) for editing a draft of this manuscript.

This study was supported by KAKENHI [grant number JP18K04416] from JSPS and the Steel Foundation for Environmental Protection Technology, Japan.

All relevant data are included in the paper or its Supplementary Information.

T.O. and K.S. have patent #Japanese Patent No. 6029081 issued to Licensee.