The current investigation reports the effect of cathode electron acceptors on simultaneous sulfide and nitrate removal in two-chamber microbial fuel cells (MFCs). Potassium permanganate and potassium ferricyanide were common cathode electron acceptors and evaluated for substrate removal and electricity generation. The abiotic MFCs produced electricity through spontaneous electrochemical oxidation of sulfide. In comparison with abiotic MFC, the biotic MFC showed better ability for simultaneous nitrate and sulfide removal along with electricity generation. Keeping external resistance of 1,000 Ω, both MFCs showed good capacities for substrate removal where nitrogen and sulfate were the main end products. The steady voltage with potassium permanganate electrodes was nearly twice that of with potassium ferricyanide. Cyclic voltammetry curves confirmed that the potassium permanganate had higher catalytic activity than potassium ferricyanide. The potassium permanganate may be a suitable choice as cathode electron acceptor for enhanced electricity generation during simultaneous treatment of sulfide and nitrate in MFCs.
Sulfide-containing waste streams are generated by many industries, such as tanneries, petrochemical plants and viscose rayon factories (Mahmood et al. 2007a; Zhang et al. 2009). Sulfide is a toxic ionic species, which exerts various toxicological impacts on human health and environmental ecology (Jin et al. 2013). The sulfide treatment may involve various physical, chemical and biological processes. Among these technologies, the biological processes are relatively cost-effective as they operate under natural ambient conditions without any requirement for expensive chemicals and catalysts (Cirne et al. 2008). Nitrate can be used to control the sulfide generation under anoxic or anaerobic conditions by some bacterial species (Garcia-de-Lomas et al. 2007). For such reasons, the simultaneous anaerobic sulfide and nitrate removal process has been recently developed.
A few researchers have combined simultaneous anaerobic sulfide and nitrate removal process in microbial fuel cells (MFCs), which is a novel approach in the field of wastewater treatment (Logan et al. 2006). The novel process can generate electricity from the biotransformations of inorganic substrates in MFCs. There are only a few published reports on the simultaneous anaerobic nitrate and sulfide removal in MFCs. Lee et al. (2012) confirmed that the MFC was capable of the simultaneous sulfide and nitrate removal using monoculture of Pseudomonas sp. C27. A two-chambered MFC was operated for simultaneous sulfide and nitrate treatment using activated sludge (Cai & Zheng 2013; Cai et al. 2013). It was concluded that the electricity output enhanced through the microbial enrichment during the operation, suggesting that an active bacterial consortium capable of substrate removal and electricity generation had established in the MFC. The effect of electrode types and operating modes on the performance of the process has been investigated by our research group (Cai et al. 2014a, b).
MATERIALS AND METHODS
Inoculum and enrichment of microbial communities
Inoculum was collected from the anaerobic methanogenic reactor operated at Dengta wastewater treatment plant located near Hangzhou City in China. Its total solids (TS) and volatile suspended solids (VSS) were 95.03 g L−1 and 68.68 g L−1, respectively, with VSS/TS ratio of 0.72. The simultaneous anaerobic sulfide and nitrate removal reactor was operated under lithoautotrophic conditions where sulfide was used as electron donor and nitrate was employed as electron acceptor. For the initial 1 month, the reactor was fed with synthetic wastewater in order to acclimatize the bacterial consortia to the new substrates and to enrich the functional bacterial populations.
The MFC was fed with synthetic influent containing NaHCO3, MgCl2 and KH2PO4 (1 g L−1 each); (NH4)2SO4 (0.24 g L−1) and trace element solution (1 mL L−1). The trace element solution was prepared according to Mahmood et al. (2007b). The nitrate-nitrogen and sulfide-sulfur concentrations were added as potassium nitrate (KNO3) and sodium sulfide (Na2S·9H2O), respectively, with their concentrations adjusted based on the type of experiment conducted.
The MFC consisted of anode and cathode chambers, and total volume of both was 350 mL (300 mL net volume) as reported in our previous study (Cai & Zheng 2013). The chambers were connected by a cation exchange membrane (CEM) (Ultrax CMI-7000 Membrane International, USA). The electrodes were graphite rods (6 cm × Φ1 cm, 18.80 cm2 net superficial area, Beijing Jixing Sheng'an Industry & Trade Co., Ltd), which were placed at the centers of each chamber and were parallel to the CEM. The external resistance was 1,000 Ω, applied to control electron flow from the anode to the cathode. Anaerobic sludge (100 mL) was inoculated in the anode chambers of biotic MFCs, and the solution in anodic chambers was circulated by a peristaltic pump. The cathode medium was 100 mg L−1 potassium permanganate (KMnO4, 50 mM PBS buffer, pH 7.0) or 625 mg L−1 potassium ferricyanide (K3[Fe(CN)]6, 50 mM PBS buffer, pH 7.0). The concentrations of cathode electron acceptors are related to the electron transfers between anode and cathode chambers. Hence, the concentrations of potassium permanganate and potassium ferricyanide which could transfer the same number of electrons were calculated. The solutions were recycled by peristaltic pumps in the cathode compartment using a 2.5 L external buffer vessel, which maintained the stability of concentration of cathode electron acceptors during the operation of MFCs.
Two MFCs of same configuration were used in the experiment. One of the MFCs used potassium permanganate as the cathode electron acceptors, while the other used potassium ferricyanide.
The two MFCs were operated in batch mode at room temperature. The synthetic wastewater was fed into the anode of the MFC on a daily basis. The final concentration of sulfide added to the MFC was 100 mg L−1 after the anodic chamber was purged with N2 for 5 min in order to remove dissolved oxygen from the solution. The nitrate concentration was increased according to stoichiometry of the chemical reaction (with S/N molar ratio of 5:2). The synthetic wastewater was fed to the anode chamber of the MFC every day. The MFC was operated under the circumstances until the effluent quality became stable, which was repeated three times at least. Subsequently, the influent substrate concentrations were increased to the next level. Three sulfide concentrations were studied: 100, 160 and 300 mg L−1.
The effluent substrate concentrations were analyzed 22 hours after the injection of the influent solution.
The physico-chemical parameters of influent and effluent, such as pH, nitrate, nitrite, sulfide and sulfate, were analyzed during the experiment. Nitrate was analyzed through the ultraviolet spectrophotometric screening method on a daily basis and nitrite was measured through the colorimetric method (APHA et al. 1998). The sulfide was determined by the iodometric method whereas sulfate was measured through the turbimetric method (APHA et al. 1998). The pH was determined following the standard method (APHA et al. 1998).
Voltage across the 1,000 Ω resistor was recorded at an interval of 10 min using a digital acquisition system (Agilent 34970A data acquisition/switch unit). A cyclic voltammeter (660C, CH Instruments Inc., USA) was used to analyze the redox status of substrates. The potential range of −2.0 V and +2.0 V was applied for this purpose. Graphite rods were used as the working electrode and platinum (CHI115, CH Instruments) as the counter electrode. Ag/AgCl electrode was also utilized as reference electrode. Voltage rate of 10 mV s−1 was chosen as scan rate in CV analysis. All tests were repeated at least three times for quality assurance.
RESULTS AND DISCUSSION
Abiotic MFCs with different cathode acceptors
Two MFCs operating as abiotic MFCs without the anaerobic sludge in anode chambers were designated as control MFCs, to evaluate the chemical sulfide and nitrate removal under anaerobic conditions. The anode chambers of both abiotic MFCs were fed with 14 mg L−1 nitrate and 60 mg L−1 sulfide, while the individual cathode electron acceptors were potassium permanganate and potassium ferricyanide. The concentrations of substrates in the anode chambers were regularly measured. Although cathode electron acceptors in the two abiotic MFCs were different, the dynamics of substrate concentrations were almost the same (data not shown). The sulfide concentration decreased to lower than 1.0 mg L−1 after 5 hours, and it was undetectable after 7 hours. The sulfate and thiosulfate concentrations slightly changed during the process. With the passage of time, the solution in the anode chambers became turbid, and white precipitates were clearly observed at the bottom of the anode chamber, which suggested that element sulfur was generated in both MFCs. However, the nitrate and nitrite concentrations in the anodic chamber did not significantly change during that period.
The maximum voltage from the abiotic MFC with potassium ferricyanide was 283 mV, generated during the first few minutes (Figure 1). Subsequently it rapidly dropped and decreased to 10 mV after 4 hours, a decrease of 96.5%. It gradually decreased to lower than 1 mV after 6 hours. The steady voltage of the abiotic MFC with potassium ferricyanide was approximately 0.1 mV. The cathode potential of the MFC still remained constant; the value was about 480 mV (vs NHE).
Evidently, power generation in abiotic MFCs was related to spontaneous electrochemical oxidation of sulfide. Sulfide is an electrochemically active oxidizable component which can directly donate electrons to the anodic electrode in MFCs. Previously, many researchers have demonstrated sulfide oxidation into elemental sulfur, which is an electrochemical reaction (Dutta et al. 2008; Gong et al. 2013; Zhang et al. 2013). Dutta et al. (2010) demonstrated that sulfide oxidation was spontaneous in an abiotic fuel cell reactor with K3Fe(CN)6 as cathode electron acceptor. It is well established that elemental sulfur is the major end product of electrochemical sulfide oxidation. A similar trend was observed in the results of the present study.
The voltage is the difference in potentials at the anode and cathode. Cathode potential is determined by characteristics of the cathode electron acceptors, while anode potential is dependent on the chemical reactions in the anode chamber. The nature of sulfide oxidation in the anode chambers was similar, implying that the distinction between the voltages outputs of the two MFCs was caused by the cathode electron acceptor. The cathode electron acceptor has obvious bearing on the voltage output of MFCs, which is directly related with redox potential of materials. In the current experiment, the redox potential of potassium permanganate was 725 mV while that of potassium ferricyanide was 480 mV. It was suggested that the MFC with potassium permanganate as cathode electron acceptor showed better capacity to generate electricity without anaerobic sludge.
Biotic MFCs with different cathode acceptors
Two MFCs with 100 mL anaerobic sludge in anode chambers were used as biotic MFCs (experimental MFCs). These were operated to demonstrate the influence of cathode electron acceptors on the performance of simultaneous anaerobic sulfide and nitrate removal under biological conditions.
Figure 2 shows that both MFCs displayed slight differences in substrate removal and both of them showed good capacity for simultaneous sulfide and nitrate removal. For the tested substrate concentration range, about 70.3–87.4% sulfide converted to sulfate and about 91.6–99.4% nitrate converted to nitrogen (Figure 2). It was suggested that nitrogen and sulfate were the main end products in both MFCs. It was inferred that the nature of biological reactions was similar in the anode chambers. However, the steady voltages of the MFC with potassium permanganate cathode electron acceptor were 2.67 times, 2.29 times and 2.04 times higher than those of the MFC with potassium ferricyanide, when the influent sulfide concentrations were 100, 160 and 300 mg/L, respectively (Figure 3). The results were considered in alignment with the previous reported studies. Employing S. putrefaciens as biocatalyst in the anode chamber with the addition of LB media as anolyte, Pandit et al. (2011) found that the MFC with potassium permanganate generated higher voltage and power density (1.11 V and 116.2 mW m−2) than that using potassium ferricyanide (0.78 V and 40.6 mW m−2). Jafary et al. (2013) used syrup with a high sugar content as a substrate in a dual-chambered MFC. When potassium ferricyanide was used as an oxidizing agent in the cathode side, the maximum output of power density was 65 mW m−2. Power density increased almost 2.5-fold and reached 234 mW m−2 when the electron acceptor in the cathode chamber was replaced with potassium permanganate.
Electrochemical behavior of abiotic and biotic MFCs
The voltammograms of abiotic MFCs using different cathode electron acceptors are shown in Figure 5(a). The voltammogram of the abiotic MFC with potassium permanganate had two oxidation–reduction peaks, which were approximately at 0.1 V (vs Ag/AgCl) and −0.9 V (vs Ag/AgCl), respectively, with the peak current of about 0.027 A. In the other abiotic MFC with potassium ferricyanide as cathode electron acceptor, the voltammogram also had two oxidation–reduction peaks, which were approximately at 0.3 V (vs Ag/AgCl) and −1.0 V (vs Ag/AgCl), respectively, and the peak current was about 0.008 A.
Voltammograms of biotic MFCs were obtained by using different cathode electron acceptors, which are shown in Figure 5(b). The voltammogram of the biotic MFC using potassium permanganate had only one oxidation–reduction peak (approximately at 0.3 V (vs Ag/AgCl)) with the peak current of about 0.029 A, while the voltammogram of MFC with potassium ferricyanide also had one oxidation–reduction peak at 0.3 V (vs Ag/AgCl) with the peak current value of 0.023 A.
Thus, abiotic MFCs and biotic MFCs both displayed electrochemical activity. However, both differed in the redox components. It was indicated that microorganisms attached at the anode had higher electrochemical performances in MFCs. Although the CVs of both biotic MFCs with different cathode electron acceptors were similar, the maximum current of MFC using potassium permanganate was higher than that with potassium ferricyanide. It suggested that potassium permanganate had better catalytic activity than potassium ferricyanide in the MFC treating sulfide and nitrate, which was consistent with the power generation.
The current investigation explored the suitability of cathode electron acceptors in MFCs treating sulfide and nitrate simultaneously. The electricity generation in abiotic MFCs was suggested to be related with spontaneous electrochemical oxidation of sulfide. The biotic simultaneous sulfide and nitrate removal in MFCs was promising and produced nitrogen and sulfate as the main end products. The steady voltage of biotic MFC using potassium permanganate was nearly twice that of with potassium ferricyanide. The CV curves supported that potassium permanganate was a better choice regardless of using anaerobic sludge. Potassium permanganate was suggested to be a more suitable cathode electron acceptor for biotic MFC simultaneously treating sulfide and nitrate. The findings are useful in actual engineering facilities involving simultaneous treatment of sulfide- and nitrate-rich wastewaters in MFCs.
The authors wish to thank the National Natural Science Foundation of China (No. 51278457) for financial support of this study.