Appl Microbiol Biotechnol (2014) 98:9879–9887 DOI 10.1007/s00253-014-6136-2
BIOENERGY AND BIOFUELS
Effect of static magnetic field on electricity production and wastewater treatment in microbial fuel cells Qinqin Tao & Shaoqi Zhou
Received: 2 September 2014 / Revised: 1 October 2014 / Accepted: 4 October 2014 / Published online: 19 October 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract The effect of a magnetic field (MF) on electricity production and wastewater treatment in two-chamber microbial fuel cells (MFCs) has been investigated. Electricity production capacity could be improved by the application of a lowintensity static MF. When a MF of 50 mT was applied to MFCs, the maximum voltage, total phosphorus (TP) removal efficiency, and chemical oxygen demand (COD) removal efficiency increased from 523±2 to 553±2 mV, ∼93 to ∼96 %, and ∼80 to >90 %, respectively, while the start-up time and coulombic efficiency decreased from 16 to 10 days and ∼50 to ∼43 %, respectively. The MF effects were immediate, reversible, and not long lasting, and negative effects on electricity generation and COD removal seemed to occur after the MF was removed. The start-up and voltage output were less affected by the MF direction. Nitrogen compounds in magnetic MFCs were nitrified more thoroughly; furthermore, a higher proportion of electrochemically inactive microorganisms were found in magnetic systems. TP was effectively removed by the co-effects of Q. Tao : S. Zhou College of Environment and Energy, South China University of Technology, Guangzhou Higher Education Mega Center, Guangzhou 510006, People’s Republic of China S. Zhou Guizhou Academy of Sciences, Shanxi Road 1, Guiyang 550001, People’s Republic of China S. Zhou State Key Laboratory of Subtropical Building Sciences, South China University of Technology, Guangzhou 510641, People’s Republic of China S. Zhou (*) Key Laboratory of Environmental Protection and Eco-remediation of Guangdong Regular Higher Education Institutions, South China University of Technology, Guangzhou Higher Education Mega Center, Guangzhou 510006, People’s Republic of China e-mail: [email protected]
microbe absorption and chemical precipitation. Chemical precipitates were analyzed by a scanning electron microscope capable of energy-dispersive spectroscopy (SEM-EDS) to be a mixture of phosphate, carbonate, and hydroxyl compounds. Keywords Microbial fuel cell (MFC) . Magnetic field (MF) . Phosphorus . Nitrogen . Electricity production . Wastewater treatment
Introduction Along with industry development and centralized urbanism across China, energy crisis and water pollution have been accelerated and triggered to sometimes extraordinary degrees. Microbial fuel cells (MFCs) have generated considerable interests among academic researchers recently as a promising technology to simultaneously treat wastewater and generate electricity (Jiang et al. 2010a, b, 2011; Butler and Nerenberg 2010). In a MFC, the electrons produced by bacteria in the anode chamber are absorbed by the anode and are transported to the electron acceptors (e.g., oxygen) in the cathode chamber through an external circuit. In regard to the application for wastewater treatment, advanced treatment such as phosphorus and nitrogen removal is important. Generally, phosphorus and nitrogen are two main contaminants of both domestic and animal wastewater, which can lead to eutrophication of water and deterioration of water quality (Domagalski et al. 2007). Precipitation is a promising method for phosphorus removal from wastewater (Cusick and Logan 2012; Nelson et al. 2003). It is well known that, due to the transport of cations through the proton exchange membrane (PEM) and the consumption of protons in the oxygen reduction reaction, the pH in the cathode chamber increases significantly. It has been reported that it can reach values close to pH 12 (Rozendal et al. 2009). Recently, some studies have
9880
shown that MFC technology could enable phosphorus recovery from animal wastewater (Ichihashi et al. 2012; Ichihashi and Hirooka 2012). Widely applied approaches for N removal utilize a two-step process consisting of autotrophic nitrification and heterotrophic denitrification. Nitrification and denitrification both may be accomplished by microorganisms in the aerobic cathode chamber (Virdis et al. 2010) and anaerobic cathode chamber (Puig et al. 2011, 2012), respectively. In the aerobic cathode chamber, ammonia nitrogen was oxidized, while in the anaerobic cathode chamber, nitrite nitrogen and nitrate nitrogen (Virdis et al. 2008, 2009, 2010) were reduced to nitrogen gas. Over the past hundred years (1911 to the present), the power generation of MFCs has grown exponentially. However, to date, the achievable power output of MFCs is still too low to make MFCs commercially feasible (Lovley 2008) and even cannot meet the power requirements for wastewater treatment facilities. A lot of research has been done to promote MFC performance through several approaches, such as optimizing the cell configuration (Liu et al. 2005), looking for high-efficiency exoelectrogens (Logan 2009) and new materials to modify the electrode (Rabaey et al. 2003), and employing a magnetic field (MF) (Li et al. 2011; Yin et al. 2013). The MF effect has been used in wastewater treatment for a few years. The MF has effects on biomass metabolism, enzyme activity, and cell membrane permeability, etc. (Liu et al. 2008). The additional MF could improve biological activity and increase the degradation rate of organic substrates in wastewater treatment systems (Yavus and Celebi 2000; Li et al. 2011). Li et al. (2011) also found that the maximum voltage increased by 20∼27 % when a MF of 100 mT was applied to the Shewanella-inoculated MFCs. Previous research has shown that MF at an appropriate intensity can also accelerate the start-up for the anammox process (Liu et al. 2008) and MFCs (Yin et al. 2013). Based on the above results, we hypothesized the possibility that MFC performances on electricity generation and wastewater treatment could be improved with the MF application. The effects of MF on voltage output, P and chemical oxygen demand (COD) removal, N removal, and transformation were investigated in this study. Furthermore, the effects of MF intensity and direction on MFCs’ start-up and voltage output were also investigated. Sediment deposited on the inner wall of the cathode chamber was characterized by scanning electron microscopy coupled with energy-dispersive spectroscopy (SEM-EDS).
Materials and methods Reactor configurations A series of two-chamber MFCs with the same structure and size were used to investigate the MF effects under different
Appl Microbiol Biotechnol (2014) 98:9879–9887
operating conditions. Two-chamber MFCs were constructed based on the tubular reactor design of Liu and Logan (2004). A two-chamber MFC consisted of two 4-cm-long cylindrical chambers (3-cm diameter, formed of Plexiglas plastic) that were separated by a PEM (3-cm diameter, Nafion 117). An anode (3-cm diameter, carbon paper) and a cathode (3-cm diameter, carbon cloth, containing 0.5 mg Pt/cm2) were placed on opposite sides of the PEM. The catholyte was kept in a 100-mL brown glass bottle and continuously recirculated into the cathode chamber using a peristaltic pump at a flow rate of approximately 20 mL/min, as previously described by Ichihashi and Hirooka (2012). Oxygenation of the solution in the blown glass bottle was supplied by an air pump, and the airflow was adjusted with an airflow rotameter. The anode and cathode were connected through an external resistor with resistance 1000 Ω. Two similar magnets were symmetrically attached to the external walls of the anode chamber and cathode chamber at a direction parallel to the anode/cathode/ PEM surface. A schematic diagram of the experimental setup is shown in Fig. 1. Start-up and operation The anode chambers and cathode chambers were inoculated with a mixture (1:3:3, v/v/v) of effluent from a parent MFC (Tao et al. 2014), anaerobic sludge, and aerobic sludge collected from the Shijing municipal wastewater treatment plant, Guangzhou, China. Two-chamber MFCs were operated in fed-batch mode with synthetic wastewater at room temperature (25∼33 °C). The synthetic wastewater mainly contained NaHCO3 5.96 g/L, NaC2H3O2 1.00 g/L, KH2PO4 0.54 g/L, NH4Cl 0.21 g/L, metals, trace minerals, and vitamins. The initial pH of synthetic wastewater was 7.00±0.12. Synthetic wastewater was fed into the anode chamber every time the voltage decreased to less than 50 mV during the stable voltage output stage. The effluent from the anode chamber was subsequently directed into the cathode chamber. The dissolved oxygen (DO) of the anolyte and catholyte was controlled at 0.03±0.01 and 3.51±0.10 mg/L, respectively, throughout the study. In the first test, seven two-chamber MFCs (MFC0, MFC1a, MFC2a, MFC3a, MFC1r, MFC2r, and MFC3r) were set up for the experiments to characterize the effects of MF intensity and direction on MFC performance. The MFCs were exposed to MFs of 0, 50, 100, 200, 50, 100, and 200 mT (the MF intensity refers to the average value at the external wall of the anode chamber or cathode chamber next to the magnet), respectively. Outside MFC1a, MFC2a, and MFC3a, two magnets facing each other were different magnetic poles (the MFs applied were called attractive MFs in this work), while outside MFC1r, MFC2r, and MFC3r, two magnets facing each other were the same magnetic poles (the MFs applied were called repellant MFs in this work). To investigate the persistency
Appl Microbiol Biotechnol (2014) 98:9879–9887
9881
Fig. 1 Schematic diagram of MFC experimental installation loaded with attractive MF (a) and repellant MF (b)
effects of MF on MFC, MFC0, MFC1a, and MFC4 were operated in the second test, with a varied MF at three stages. A MF was applied to MFC1a at stages 1 and 3 and to MFC4 at stage 2. MFC0 without applying MF was still run as the control throughout the second test.
examine the morphology of the samples as well as elemental composition.
Results Analysis
Effects of MF intensity and direction on MFC start-up
The output voltages across the external resistances of the MFCs were automatically recorded at 1-min intervals using a data acquisition system (M2700, Keithley, USA) connected with a computer. Current (I) and power (P=IV) were calculated according to Ohm’s law. The areal current density and power density were calculated by dividing current and power by the net anode or cathode area (7.065 cm2). Coulombic efficiencies (CE) were calculated as previously described (Chen et al. 2012). Polarization curves were detected by verifying external resistances from 30 to 50,000 Ω with an interval of 1 min to gain stable voltages. The internal resistances and maximum power densities were obtained by analyzing the polarization curves (Watanabe 2008). Water samples from the anode chamber and cathode chamber were taken at the end of an electricity production cycle. Water samples were centrifuged at 8000 rpm for 5 min, and the supernatants were used for measurement. Chemical oxygen demand (COD), total nitrogen (TN, the sum of ammonium (N-NH4+), nitrates (N-NO3−), and nitrites (N-NO2−)), and total phosphorus (TP) were measured according to the standard methods (APHA) (APHA 2005). Values of pH were measured using a pH meter (PHS-25, Leici, China). Solution DO values were measured using a dissolved oxygen meter (JPB-607, Leici, China). Precipitates formed in the cathode chambers were washed and re-suspended in Milli-Q water to remove soluble chemicals, and then they were centrifuged again at 8000 rpm for 5 min. These re-suspension and centrifugation steps were repeated twice, and finally, the precipitates were dried in a dried pot. The precipitates were analyzed by scanning electron microscopy coupled with energy-dispersive spectroscopy (SEM-EDS, Carl Zeiss EVO LS10, UK) to
Figure 2 shows the voltage outputs’ change from start to stable operation of the MFCs under different MF intensities and directions. The red arrows represent the first time for MFCs to reach the maximum power production. Cell voltages over subsequent cycles were then reproducible in terms of maximum voltage. When the MFCs were exposed to attractive MFs (Fig. 2a), the time required to reach the first maximum power cycle was significantly shorter than the control. MFC1a with an attractive MF of 50 mT started fastest; it required 10 days before reaching maximum power production. The maximum voltage under these conditions was 553±2 mV. For MFC2a and MFC3a with attractive MFs of 100 and 200 mT, respectively, the times required to reach the first maximum power cycle both increased to ∼13 days, and the maximum voltages were 536±2 and 517±1 mV, respectively. The time required for MFC0 to reach the first maximum power cycle was 16 days, which was the longest, and the maximum voltage was 523±2 mV. These results suggest that the attractive MFs accelerated the start-up. In addition, compared to the control (MFC0), a slight increase in the maximum output voltage was observed for MFC1a and MFC2a. However, the maximum output voltage for MFC3a did not increase but decreased a little. The results indicate that the attractive MF application is of benefit to electricity production, but the intensity of the applied MF should not be too high; otherwise, it would have negative effects on electricity production. To further explore the effects of MF direction on the startup and electricity production of MFCs, repellant MFs were applied to MFC1r, MFC2r, and MFC3r. Figure 2b shows the voltage outputs’ change from start to stable operation of the
9882
Appl Microbiol Biotechnol (2014) 98:9879–9887 A
B
0.6
0.5 MFC0
0.4
Voltage(V)
Voltage(V)
0.6
MFC1a
0.5 MFC0
0.3 0.2 0.1
0.5 MFC2r
MFC3a
Voltage(V)
0.4 0.3 0.2
0.3 0.2
0.0
5000 10000 15000 20000 250000
5000 10000 15000 20000 25000
Time (min)
Time (min)
MFCs under repellant MFs. No distinct difference was found between MFC1a and MFC1r, MFC2a and MFC2r, and MFC3a and MFC3r in terms of start-up time and maximum voltage. This suggests that the MF direction exhibited negligible influence on start-up and voltage output in this work. Phosphorus and nitrogen removal and crystal accumulation In this study, the effect of MF on phosphorus removal, nitrogen removal, and transformation was investigated. Concentrations of TP, N-NH4+, N-NO3−, and N-NO2− in the effluent from the anode chamber and cathode chamber of MFC0, MFC1a, MFC2a, and MFC3a were measured, and results are shown in Table 1. Nitrogen and phosphorus in the synthetic wastewater were in the form of N-NH4+ and P-PO43− in the beginning. The concentrations of N-NH4+ and P-PO43− were 55.0 and 122.9 mg/L, respectively. The concentrations of TP, NNO3−, and N-NO2− in the effluent of anode chambers were nearly the same as those of the synthetic water. This is probably due to the low DO in the anolytes, which led to oxygen shortage for bacteria in anode chambers aerobic phosphorus uptake and ammonia nitrogen nitrification or nitrosification. However, the concentration of TN in the effluent of anode Table 1 TP, N-NH4+, N-NO3−, and N-NO2− concentrations in the effluents from MFCs under different attractive MF intensities
Effluent of cathode chamber
MFC3r
0.4
0.1
0
Effluent of anode chamber
0.2
0.6
0.5 MFC2a
0.0
MFC0 MFC1a MFC2a MFC3a MFC0 MFC1a MFC2a MFC3a
0.3
0.0
0.6
0.1
Synthetic water
MFC1r
0.4
0.1
0.0
Voltage(V)
Fig. 2 MFC voltage outputs change during the start-up period. a MFC1a, MFC2a, and MFC3a under attractive MFs of 50, 100, and 200 mT, respectively; b MFC1r, MFC2r, and MFC3r under repellant MFs of 50, 100, and 200 mT, respectively. MFC0 without applying MF was run as the control throughout the study
TP (mg/L)
N-NH4+ (mg/L)
N-NO3− (mg/L)
N-NO2− (mg/L)
122.9 120.7±2.5 122.1±2.6 121.3±1.8 120.5±3.3 8.4±1.7 4.8±0.3 5.1±0.7 5.7±0.4
55.0 44.8±1.2 40.2±1.5 41.9±0.7 43.4±2.0 5.2±0.9 3.6±0.8 4.1±0.5 4.7±1.1
0 ≈0 2.6±1.0 2.2±0.8 2.5±0.7 1.3±0.5 37.2±1.6 36.1±2.3 35.8±2.4
0 ≈0 ≈0 ≈0 ≈0 34.8±2.7 0.5±0.3 0.4±0.2 0.7±0.1
0
5000 10000 15000 20000 250000 Time (min)
5000 10000 15000 20000 25000 Time (min)
chambers decreased about 10 mg/L compared with that of synthetic wastewater. The decrease of TN may be due to the consumption for microorganisms’ growth and reproduction, ammonium ion diffusion from the anode chamber to the cathode chamber, or ammonium volatilization. In comparison with MFC0, there was no significant difference in the concentrations of TP, N-NH4+, and N-NO2− in the effluent from anode chambers except a small quantity of N-NO3− presented in anode chambers of magnetic MFCs. TP removal efficiencies in MFC0, MFC1a, MFC2a, and MFC3a were ∼93, ∼96, ∼96, and ∼95 %, respectively. Compared with the control, there was a slight increase in TP removal as a MF was applied to MFCs. Some sediment accumulated on the inner walls of cathode chambers after operating for a period of time. Sediment was characterized by SEM-EDS. The results of MFC0 and MFC1a are shown in Figs. 3 and 4 (results of MFC2a and MFC3a were not shown for their similarity). SEM images show that sediment in cathode chambers was a mixture of microbe cells and chemical precipitates. The EDS patterns from the chemical precipitates in the cathode chambers of MFC0 and MFC1a are similar. The chemical precipitates were analyzed to be a mixture of phosphate, carbonate, and hydroxyl compound in an earlier study (Tao et al. 2014). Chemical precipitates formed in cathode chambers most likely resulted from the progressively alkaline catholyte. The pH values of the effluent from cathode chambers of MFC0, MFC1a, MFC2a, and MFC3a were 9.27 ±0.11, 9.38±0.13, 9.35±0.15, and 9.33±0.14, respectively. The slight increase in TP removal may be attributed to the higher pH in the cathode chamber of magnetic MFCs. The precipitation of phosphorus could be promoted in the presence of MF. TP was probably removed by microbe absorption and chemical precipitation, and the latter may be the major factor (Tao et al. 2014). Though there was almost no TN removal in all cathode chambers, the rate of N-NH4+ converted to N-NO3− or NNO2− was very different. In the control, a large proportion of N-NH4+ was converted to N-NO2−. In magnetic MFCs, however, the vast majority of N-NH4+ was converted to N-NO3−. The results show that there does exist the possibility of partial
Appl Microbiol Biotechnol (2014) 98:9879–9887
9883
Fig. 3 SEM images of sediment deposit on the inner walls of the cathode chamber of MFC0 (a) and MFC1a (b)
nitrification in the cathode chamber of MFC0 but more thorough nitrification in the cathode chambers of magnetic MFCs. Substrate degradation and coulombic efficiency COD removal efficiencies and coulombic efficiencies of magnetic MFCs and control were calculated. As shown in Fig. 5, considerably higher COD removal efficiencies were witnessed in the magnetic MFCs compared to the control (∼80 %). Moreover, the highest COD removal efficiency (>90 %) was obtained in MFC1a (under a MF of 50 mT). Despite the better degradation performance, the coulombic efficiencies of the magnetic MFCs were relatively lower than those of the control. The results indicated magnetic systems with a higher percentage of electrochemically inactive microorganisms. Polarization and power density curves The polarization curves of the magnetic MFCs and the control are illustrated in Fig. 6. The maximum power densities obtained in MFC0, MFC1a, MFC2a, and MFC3a were 526, 548, 539, and 523 mW/m2, respectively. The internal resistances were 232, 207, 221, and 241 Ω, respectively. The results imply that low-intensity MF (e.g., 50 and 100 mT) could reduce the MFC internal resistance and then improve the maximum power density. High-intensity MF may have negative effects Fig. 4 Energy-dispersive spectrographs of precipitate in SEM image a and b
on internal resistance and power density. The polarization and power density curves also demonstrate a decrease in activation losses under a MF at an appropriate intensity (50 or 100 mT in this work). It implies that the metabolic and electrochemical activities of some microorganisms were likely to be improved under MF exposure, which lowered the activation loss. This conclusion was supported by the higher COD removal in magnetic MFCs. Persistency effects of MF on MFC The persistency effects of MF on electricity generation and TP, N-NH4+, N-NO3−, N-NO2−, and COD removal were explored in this work. The changes of the output voltage of magnetic MFCs with varied MFs and control are demonstrated in Fig. 7a. MFC1a and MFC4 were operated under MFs of 50 and 0 mT at the first stage. In comparison with MFC0 and MFC4, a slight increase in the maximum output voltage was observed for MFC1a at stage 1. For MFC4, a sudden increase (though the increase was modest) of voltage was observed when a MF was applied at stage 2, but the voltage dropped sharply to less than 50 mV at stage 3 when the MF was removed. For MFC1a, the voltage suddenly dropped sharply to less than 50 mT when the MF was removed at stage 2, but the voltage rose again to the original level when resuming the MF at stage 3. The immediate increase in voltage for MFC4 after employing a MF at stage 2 and for MFC1a after
B
A O Mg
P
O
Mg
P
C Na
0
Mn
Mn
K
C Na
5
0
5
9884
Appl Microbiol Biotechnol (2014) 98:9879–9887
100
100
80 COD removal efficiency Coulombic efficiency
60
60
40
40
20
20
0
Coulombic efficiency (%)
COD removal efficiency (%)
80
(the results are in Table 1). The residual TP, N-NH4+, N-NO3−, and N-NO2− in the effluent from the cathode chamber of MFC4 and MFC1a were at the same levels as that in the control when not employing a MF to MFC1a at stage 2 and to MFC4 at stages 1 and 3. However, diversified COD removal efficiency was observed for MFC4 and MFC1a with a varied MF at three stages. Higher COD removal efficiency was obtained after employing a MF for MFC4 at stage 2 and for MFC1a at stages 1 and 3, but dropped sharply to lower than that in the control for MFC 4 at stage 3 and for MFC1a at stage 2 when the MF was removed. The above results indicate that the effect of MF on wastewater treatment in MFCs was also immediate and reversible. The higher COD removal efficiencies for MFC4 at stage 2 and for MFC1a at stages 1 and 3 show that MF plays a positive role in promoting COD removal. However, the drop of COD removal efficiencies to lower than that in the control for MFC 4 at stage 3 and for MFC1a at stage 2 indicates that there may be some negative effects when the MF was removed. Approximately 70 % COD was removed in MFC1a at stage 2 and in MFC4 at stage 3; at this moment, however, the voltage of MFC1a and MFC4 was less than 50 mV; this suggests that the great mass of COD was removed by electrochemically inactive microorganisms.
0 0
100 50 Magnetic field strength (mT)
200
Fig. 5 COD removal efficiency and coulombic efficiency of MFC under different attractive MF intensities
resuming a MF at stage 3 indicates that the enhancement effect of MF seems to be immediate and reversible. In addition, the rapid recovery of electricity production after resuming the MF suggests that the MF may promote the electron transfer or catalytic activity of electro-microorganisms rather than causing irreversible changes to the cell structure. The sharp decline of voltage after removing the MF for MFC4 at stage 3 and for MFC1a at stage 2 implies that the positive effect of MF on electricity production seemed to be nonpersistent and even has negative effects when suddenly pausing the MF, possibly leading to a period of shutdown for the microorganisms that have adapted to the MF condition. The sharp decline of voltage after MF removal should be mainly caused by a pause of MF rather than the substrate depletion. This was supported by the relatively high voltage in the control. In order to validate the long-term effect of MF on wastewater treatment, the residual TP, N-NH4+, N-NO3−, and NNO2− and COD removal efficiency in the effluent from cathode chambers were investigated. The results are shown in Fig. 7b. As seen from the results in Fig. 7b, when a MF was applied to MFC4 at stage 2 and to MFC1a at stages 1 and 3, the residual TP, N-NH4+, N-NO3−, and N-NO2− in the effluent from cathode chambers of MFC4 and MFC1a were similar to those of MFC1a with a constant MF of 50 mT in the first test
The magnetic MFCs started faster than the control under MF exposure, with a difference that varied within 10–13 days, depending on the MF intensity applied (Fig. 2). Moreover, enhancement of electricity production was observed for MFC1a and MFC2a but not for MFC3a. All of these results demonstrated that MF application could accelerate start-up and promote electricity production of MFCs. It was probably because the MFs promoted the enrichment of anode electrogenesis and accelerated the rate of proliferation. The applied MF may reduce anode activation loss, which results from microbial metabolic activity and electrochemical reactions
A
B 700
600
MFC0 MFC1a MFC2a MFC3a
MFC0 MFC1a MFC2a MFC3a
500
Power density (mW/m )
600
2
500
Voltage (mV)
Fig. 6 Polarization (a) and power density (b) curves of MFC under different attractive MF intensities
Discussion
400 300 200 100
400
300
200
100
0
500
1000
1500
2000
2500 2
Current density (mA/m )
3000
3500
0
500
1000
1500
2000
2500 2
Current density (mA/m )
3000
3500
Appl Microbiol Biotechnol (2014) 98:9879–9887
A
Stage1 0.6
Voltage (V)
Fig. 7 Variations of output voltage (a), residual TP, N-NH4+, N-NO3−, N-NO2−, and COD removal efficiency in the effluent from cathode chambers (b) of magnetic MFCs with varied MFs and control
9885
Stage3 without a MF
MFC0
0.2 0.0 without a MF
apply a MF
remove the MF
MFC4
0.4 0.2 0.0 apply a MF
0.6
Voltage (V)
Stage2 without a MF
0.4
0.6
Voltage (V)
without a MF
remove the MF
re-apply a MF
MFC1a 0.4 0.2 0.0 0
2500
5000
7500
10000
12500
15000
17500
20000
Time (min)
concentration (mg/L)
+ N-NH4
TP
Stage3
N-NO2
COD
80 60 40 20 0 80 60 40 20 0
COD removal efficiency (%)
40 30 MFC0 20 10 0 40 30 MFC4 20 10 0 40 30 MFC1a 20 10 0 2500
N-NO3
COD removal efficiency (%)
concentration (mg/L)
Stage2
Stage1
COD removal efficiency (%)
concentration (mg/L)
B
80 60 40 20 0 5000
7500
10000
12500
15000
17500
20000
Time (min)
(Yin et al. 2013). However, the intensity of the applied MF should not be too high, or it may have less effect and even negative effects on start-up and electricity production. Nitrogen compound transformations in magnetic MFCs and control were very different. Results in Table 1 show that the vast majority of residual TN in the effluent from cathode chambers of magnetic MFCs was N-NO3−, but in the control, it was N-NO2−. Nitrification is a two-step process: (1) NH4+ or
ammonia (NH3) is oxidized into NO2− by ammonia-oxidizing bacteria (AOB), often Nitrosomonas spp, and (2) the NO2− is further oxidized into NO3− by nitrite-oxidizing bacteria (NOB), often Nitrobacter spp. The nitrification process was stopped at the end of the first step maybe because of the shortage of oxygen in the control. Nitrification was carried out in the magnetic MFCs in a more complete way. This was consistent with previous research of Tomska and Wolny
9886
(2008), who found that more intensive nitrogen compound transformations and higher nitrification rate could be obtained in magnetic-activated sludge units. In addition, most of this was due to the higher oxygen uptake rate in the second nitrification phase. Many studies have suggested that MF is an intensifying factor for organic substrate degradation (Jung and Sofer 1997; Jung et al. 1993). The study of Li et al. (2011) has also shown that the MF could stimulate cell growth. One potential explanation for such improvement of microbial activity by MF is that the MF causes an oxidative stress and increases the activity, lifetime, and concentration of free radicals (Zhang et al. 2003; Jones et al. 2007). In consequence, highly reactive byproducts of usual metabolism and immune defense such as oxidative metabolites, free radicals, and reactive oxygen species (Zhang et al. 2003) increase and cause a change in enzymatic activity. Therefore, the possible mechanism could be involved with the change of bioelectrochemical reaction rate upon MF application to MFCs. A further exploration of the persistency effects of MF was carried out by applying a varied MF to the MFCs. Results in previous sections suggested that the MF effects on MFC performance for electricity generation and wastewater treatment were immediate, reversible, and not long lasting. Negative effects on electricity generation and COD removal would occur when the MF is suddenly paused. That may be because the biological behaviors of some microorganisms in magnetic MFCs have been altered; then, they will not adapt to the living environment without employing a MF. There was no adverse effect on TP removal, nitrogen removal, and transformation but on electricity generation. That means that microorganisms that take part in TP removal, nitrogen removal, and transformation in MFCs contributed little to electricity generation in this work. This conclusion was consistent with the recent research of ours. Previous research has shown that there was no nitrifying bacteria or anaerobic ammonia oxidation bacteria in the anode chamber for the long time anaerobic environment in the anode chamber. When the cathodic DO was at 3.5 mg/L, N-NO3− and N-NO2− in the catholyte would not be used as the electron acceptors for electricity generation (Tao et al. 2014). Acknowledgments Financial supports from the National Natural Science Foundation (21277052), the Department of Guangdong Education and the Science and Technology Bureau (2010), the State Key Laboratory of Subtropical Building Science (2013ZC03, 2014ZB04), and the Environmental Protection Bureau (201203) are gratefully acknowledged.
References APHA (2005) Standard methods for the examination of water and wastewater, 19th edn. American Public Health Association, Washington
Appl Microbiol Biotechnol (2014) 98:9879–9887 Butler CS, Nerenberg R (2010) Performance and microbial ecology of air-cathode microbial fuel cells with layered electrode assemblies. Appl Microbiol Biotechnol 86:1399–1408 Chen SS, Liu GL, Zhang RD, Qin BY, Luo Y, Hou YP (2012) Improved performance of the microbial electrolysis desalination and chemicalproduction cell using the stack structure. Bioresour Technol 116: 507–511 Cusick RD, Logan BE (2012) Phosphate recovery as struvite within a single chamber microbial electrolysis cell. Bioresour Technol 107: 110–115 Domagalski J, Lin C, Luo Y, Kang J, Wang S, Brown LR, Munn MD (2007) Eutrophication study at the Panjiakou-Daheiting reservoir system, northern Hebei province, People’s Republic of China: chlorophyll-A model and sources of phosphorus and nitrogen. Agr Water Manag 94(1–3):43–53 Ichihashi O, Hirooka K (2012) Removal and recovery of phosphorus as struvite from swine wastewater using microbial fuel cell. Bioresour Technol 114:303–307 Ichihashi O, Yamamoto N, Hirooka K (2012) Power generation by and microbial community structure in microbial fuel cell treating animal wastewater. J Jpn Soc Water Environ 35(1):19–26 Jiang D, Li B, Jia W, Lei Y (2010a) Effect of inoculum types on bacterial adhesion and power production in microbial fuel cells. Appl Biochem Biotechnol 160(1):182–196 Jiang D, Li X, Raymond D, Mooradain J, Li B (2010b) Power recovery with multi-anode/cathode microbial fuel cells suitable for future large-scale applications. Int J Hydrog Energy 35(16): 8683–8689 Jiang D, Curtis M, Troop E, Scheible K, McGrath J, Boxun H, Suib S, Raymond D, Baikun L (2011) A pilot-scale study on utilizing multianode/cathode microbial fuel cells (MAC MFCs) to enhance the power production in wastewater treatment. Int J Hydrog Energy 36(1):876–884 Jones AR, Hay S, Woodward JR, Scrutton NS (2007) Magnetic field effect studies indicate reduced geminate recombination of the radical pair in substrate-bound adenosylcobalamin-dependent ethanolamine ammonia lyase. J Am Chem Soc 129(50):15718–15727 Jung JT, Sofer S (1997) Enhancement of phenol biodegradation by south magnetic field exposure. J Chem Technol Biotechnol 70(3):229– 303 Jung JT, Sanji B, Godbole S, Sofer S (1993) Biodegradation of phenol: a comparative study with and without applying magnetic fields. J Chem Technol Biotechnol 56(1):73–76 Li WW, Sheng GP, Liu XW, Cai PJ, Sun M, Xiao X, Wang YK, Tong ZH, Dong F, Yu HQ (2011) Impact of a static magnetic field on the electricity production of Shewanella-inoculated microbial fuel cells. Biosens Bioelectron 26(10):3987–3992 Liu H, Logan BE (2004) Electricity generation using an air-cathode single chamber microbial fuel cell in the presence and absence of a proton exchange membrane. Environ Sci Technol 38(14):4040– 4046 Liu H, Cheng SA, Logan BE (2005) Power generation in fed-batch microbial fuel cells as a function of ionic strength, temperature, and reactor configuration. Environ Sci Technol 39(14): 5488–5493 Liu ST, Yang FL, Meng FG, Chen HH, Gong Z (2008) Enhanced anammox consortium activity for nitrogen removal: impacts of static magnetic field. J Biotechnol 138(3–4):96–102 Logan BE (2009) Exoelectrogenic bacteria that power microbial fuel cells. Nat Rev Microbiol 7(5):375–381 Lovley DR (2008) The microbe electric: conversion of organic matter to electricity. Curr Opin Biotechnol 19(6):564–571 Nelson NO, Mikkelsen RL, Hesterberg DL (2003) Struvite precipitation in anaerobic swine lagoon liquid: effect of pH and Mg: P ratio and determination of rate constant. Bioresour Technol 89(3):229–236
Appl Microbiol Biotechnol (2014) 98:9879–9887 Puig S, Serra M, Vilar-Sanz A, Cabré M, Bañeras L, Colprim J, Balaguer MD (2011) Autotrophic nitrite removal in the cathode of microbial fuel cells. Bioresour Technol 102(6):4462–4467 Puig S, Coma M, Desloover J, Boon N, Colprim J, Balaguer MD (2012) Autotrophic denitrification in microbial fuel cells treating low ionic strength waters. Environ Sci Technol 46:2309–2315 Rabaey K, Lissens G, Siciliano SD, Verstraete W (2003) A microbial fuel cell capable of converting glucose to electricity at high rate and efficiency. Biotechnol Lett 25(18):1531–1535 Rozendal RA, Leone E, Keller J, Rabaey K (2009) Efficient hydrogen peroxide generation from organic matter in a bioelectrochemical system. Electrochem Commun 11(9):1752–1755 Tao QQ, Luo JJ, Zhou J, Zhou SQ, Liu GL, Zhang RD (2014) Effect of dissolved oxygen on nitrogen and phosphorus removal and electricity production in microbial fuel cell. Bioresour Technol 164:402–407 Tomska A, Wolny L (2008) Enhancement of biological wastewater treatment by magnetic field exposure. Desalination 222(1–3): 368–373 Virdis B, Rabaey K, Yuan Z, Keller J (2008) Microbial fuel cells for simultaneous carbon and nitrogen removal. Water Res 42(12):3013– 3024
9887 Virdis B, Rabaey K, Yuan Z, Rozendal RA, Keller J (2009) Electron fluxes in a microbial fuel cell performing carbon and nitrogen removal. Environ Sci Technol 43(13):5144–5149 Virdis B, Rabaey K, Rozendal RA, Yuan Z, Keller J (2010) Simultaneous nitrification, denitrification and carbon removal in microbial fuel cells. Water Res 44(9):2970–2980 Watanabe K (2008) Recent developments in microbial fuel cell technologies for sustainable bioenergy. J Biosci Bioeng 106(6): 528–536 Yavus H, Celebi SS (2000) Effects of magnetic field on activity of activated sludge in wastewater treatment. Enzym Microbiol Technol 26(1):22–27 Yin Y, Huang GT, Tong YR, Liu YD, Zhang LH (2013) Electricity production and electrochemical impedance modeling of microbial fuel cells under static magnetic field. J Power Sources 237: 58–63 Zhang QM, Tokiwa M, Doi T, Nakahara T, Chang P, Nakamura N, Hori M, Miyakoshi J, Yonei S (2003) Strong static magnetic field and the induction of mutations through elevated production of reactive oxygen species in Escherichia coli soxR. Int J Radiat Biol 79(4): 281–286
BIOENERGY AND BIOFUELS
Effect of static magnetic field on electricity production and wastewater treatment in microbial fuel cells Qinqin Tao & Shaoqi Zhou
Received: 2 September 2014 / Revised: 1 October 2014 / Accepted: 4 October 2014 / Published online: 19 October 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract The effect of a magnetic field (MF) on electricity production and wastewater treatment in two-chamber microbial fuel cells (MFCs) has been investigated. Electricity production capacity could be improved by the application of a lowintensity static MF. When a MF of 50 mT was applied to MFCs, the maximum voltage, total phosphorus (TP) removal efficiency, and chemical oxygen demand (COD) removal efficiency increased from 523±2 to 553±2 mV, ∼93 to ∼96 %, and ∼80 to >90 %, respectively, while the start-up time and coulombic efficiency decreased from 16 to 10 days and ∼50 to ∼43 %, respectively. The MF effects were immediate, reversible, and not long lasting, and negative effects on electricity generation and COD removal seemed to occur after the MF was removed. The start-up and voltage output were less affected by the MF direction. Nitrogen compounds in magnetic MFCs were nitrified more thoroughly; furthermore, a higher proportion of electrochemically inactive microorganisms were found in magnetic systems. TP was effectively removed by the co-effects of Q. Tao : S. Zhou College of Environment and Energy, South China University of Technology, Guangzhou Higher Education Mega Center, Guangzhou 510006, People’s Republic of China S. Zhou Guizhou Academy of Sciences, Shanxi Road 1, Guiyang 550001, People’s Republic of China S. Zhou State Key Laboratory of Subtropical Building Sciences, South China University of Technology, Guangzhou 510641, People’s Republic of China S. Zhou (*) Key Laboratory of Environmental Protection and Eco-remediation of Guangdong Regular Higher Education Institutions, South China University of Technology, Guangzhou Higher Education Mega Center, Guangzhou 510006, People’s Republic of China e-mail: [email protected]
microbe absorption and chemical precipitation. Chemical precipitates were analyzed by a scanning electron microscope capable of energy-dispersive spectroscopy (SEM-EDS) to be a mixture of phosphate, carbonate, and hydroxyl compounds. Keywords Microbial fuel cell (MFC) . Magnetic field (MF) . Phosphorus . Nitrogen . Electricity production . Wastewater treatment
Introduction Along with industry development and centralized urbanism across China, energy crisis and water pollution have been accelerated and triggered to sometimes extraordinary degrees. Microbial fuel cells (MFCs) have generated considerable interests among academic researchers recently as a promising technology to simultaneously treat wastewater and generate electricity (Jiang et al. 2010a, b, 2011; Butler and Nerenberg 2010). In a MFC, the electrons produced by bacteria in the anode chamber are absorbed by the anode and are transported to the electron acceptors (e.g., oxygen) in the cathode chamber through an external circuit. In regard to the application for wastewater treatment, advanced treatment such as phosphorus and nitrogen removal is important. Generally, phosphorus and nitrogen are two main contaminants of both domestic and animal wastewater, which can lead to eutrophication of water and deterioration of water quality (Domagalski et al. 2007). Precipitation is a promising method for phosphorus removal from wastewater (Cusick and Logan 2012; Nelson et al. 2003). It is well known that, due to the transport of cations through the proton exchange membrane (PEM) and the consumption of protons in the oxygen reduction reaction, the pH in the cathode chamber increases significantly. It has been reported that it can reach values close to pH 12 (Rozendal et al. 2009). Recently, some studies have
9880
shown that MFC technology could enable phosphorus recovery from animal wastewater (Ichihashi et al. 2012; Ichihashi and Hirooka 2012). Widely applied approaches for N removal utilize a two-step process consisting of autotrophic nitrification and heterotrophic denitrification. Nitrification and denitrification both may be accomplished by microorganisms in the aerobic cathode chamber (Virdis et al. 2010) and anaerobic cathode chamber (Puig et al. 2011, 2012), respectively. In the aerobic cathode chamber, ammonia nitrogen was oxidized, while in the anaerobic cathode chamber, nitrite nitrogen and nitrate nitrogen (Virdis et al. 2008, 2009, 2010) were reduced to nitrogen gas. Over the past hundred years (1911 to the present), the power generation of MFCs has grown exponentially. However, to date, the achievable power output of MFCs is still too low to make MFCs commercially feasible (Lovley 2008) and even cannot meet the power requirements for wastewater treatment facilities. A lot of research has been done to promote MFC performance through several approaches, such as optimizing the cell configuration (Liu et al. 2005), looking for high-efficiency exoelectrogens (Logan 2009) and new materials to modify the electrode (Rabaey et al. 2003), and employing a magnetic field (MF) (Li et al. 2011; Yin et al. 2013). The MF effect has been used in wastewater treatment for a few years. The MF has effects on biomass metabolism, enzyme activity, and cell membrane permeability, etc. (Liu et al. 2008). The additional MF could improve biological activity and increase the degradation rate of organic substrates in wastewater treatment systems (Yavus and Celebi 2000; Li et al. 2011). Li et al. (2011) also found that the maximum voltage increased by 20∼27 % when a MF of 100 mT was applied to the Shewanella-inoculated MFCs. Previous research has shown that MF at an appropriate intensity can also accelerate the start-up for the anammox process (Liu et al. 2008) and MFCs (Yin et al. 2013). Based on the above results, we hypothesized the possibility that MFC performances on electricity generation and wastewater treatment could be improved with the MF application. The effects of MF on voltage output, P and chemical oxygen demand (COD) removal, N removal, and transformation were investigated in this study. Furthermore, the effects of MF intensity and direction on MFCs’ start-up and voltage output were also investigated. Sediment deposited on the inner wall of the cathode chamber was characterized by scanning electron microscopy coupled with energy-dispersive spectroscopy (SEM-EDS).
Materials and methods Reactor configurations A series of two-chamber MFCs with the same structure and size were used to investigate the MF effects under different
Appl Microbiol Biotechnol (2014) 98:9879–9887
operating conditions. Two-chamber MFCs were constructed based on the tubular reactor design of Liu and Logan (2004). A two-chamber MFC consisted of two 4-cm-long cylindrical chambers (3-cm diameter, formed of Plexiglas plastic) that were separated by a PEM (3-cm diameter, Nafion 117). An anode (3-cm diameter, carbon paper) and a cathode (3-cm diameter, carbon cloth, containing 0.5 mg Pt/cm2) were placed on opposite sides of the PEM. The catholyte was kept in a 100-mL brown glass bottle and continuously recirculated into the cathode chamber using a peristaltic pump at a flow rate of approximately 20 mL/min, as previously described by Ichihashi and Hirooka (2012). Oxygenation of the solution in the blown glass bottle was supplied by an air pump, and the airflow was adjusted with an airflow rotameter. The anode and cathode were connected through an external resistor with resistance 1000 Ω. Two similar magnets were symmetrically attached to the external walls of the anode chamber and cathode chamber at a direction parallel to the anode/cathode/ PEM surface. A schematic diagram of the experimental setup is shown in Fig. 1. Start-up and operation The anode chambers and cathode chambers were inoculated with a mixture (1:3:3, v/v/v) of effluent from a parent MFC (Tao et al. 2014), anaerobic sludge, and aerobic sludge collected from the Shijing municipal wastewater treatment plant, Guangzhou, China. Two-chamber MFCs were operated in fed-batch mode with synthetic wastewater at room temperature (25∼33 °C). The synthetic wastewater mainly contained NaHCO3 5.96 g/L, NaC2H3O2 1.00 g/L, KH2PO4 0.54 g/L, NH4Cl 0.21 g/L, metals, trace minerals, and vitamins. The initial pH of synthetic wastewater was 7.00±0.12. Synthetic wastewater was fed into the anode chamber every time the voltage decreased to less than 50 mV during the stable voltage output stage. The effluent from the anode chamber was subsequently directed into the cathode chamber. The dissolved oxygen (DO) of the anolyte and catholyte was controlled at 0.03±0.01 and 3.51±0.10 mg/L, respectively, throughout the study. In the first test, seven two-chamber MFCs (MFC0, MFC1a, MFC2a, MFC3a, MFC1r, MFC2r, and MFC3r) were set up for the experiments to characterize the effects of MF intensity and direction on MFC performance. The MFCs were exposed to MFs of 0, 50, 100, 200, 50, 100, and 200 mT (the MF intensity refers to the average value at the external wall of the anode chamber or cathode chamber next to the magnet), respectively. Outside MFC1a, MFC2a, and MFC3a, two magnets facing each other were different magnetic poles (the MFs applied were called attractive MFs in this work), while outside MFC1r, MFC2r, and MFC3r, two magnets facing each other were the same magnetic poles (the MFs applied were called repellant MFs in this work). To investigate the persistency
Appl Microbiol Biotechnol (2014) 98:9879–9887
9881
Fig. 1 Schematic diagram of MFC experimental installation loaded with attractive MF (a) and repellant MF (b)
effects of MF on MFC, MFC0, MFC1a, and MFC4 were operated in the second test, with a varied MF at three stages. A MF was applied to MFC1a at stages 1 and 3 and to MFC4 at stage 2. MFC0 without applying MF was still run as the control throughout the second test.
examine the morphology of the samples as well as elemental composition.
Results Analysis
Effects of MF intensity and direction on MFC start-up
The output voltages across the external resistances of the MFCs were automatically recorded at 1-min intervals using a data acquisition system (M2700, Keithley, USA) connected with a computer. Current (I) and power (P=IV) were calculated according to Ohm’s law. The areal current density and power density were calculated by dividing current and power by the net anode or cathode area (7.065 cm2). Coulombic efficiencies (CE) were calculated as previously described (Chen et al. 2012). Polarization curves were detected by verifying external resistances from 30 to 50,000 Ω with an interval of 1 min to gain stable voltages. The internal resistances and maximum power densities were obtained by analyzing the polarization curves (Watanabe 2008). Water samples from the anode chamber and cathode chamber were taken at the end of an electricity production cycle. Water samples were centrifuged at 8000 rpm for 5 min, and the supernatants were used for measurement. Chemical oxygen demand (COD), total nitrogen (TN, the sum of ammonium (N-NH4+), nitrates (N-NO3−), and nitrites (N-NO2−)), and total phosphorus (TP) were measured according to the standard methods (APHA) (APHA 2005). Values of pH were measured using a pH meter (PHS-25, Leici, China). Solution DO values were measured using a dissolved oxygen meter (JPB-607, Leici, China). Precipitates formed in the cathode chambers were washed and re-suspended in Milli-Q water to remove soluble chemicals, and then they were centrifuged again at 8000 rpm for 5 min. These re-suspension and centrifugation steps were repeated twice, and finally, the precipitates were dried in a dried pot. The precipitates were analyzed by scanning electron microscopy coupled with energy-dispersive spectroscopy (SEM-EDS, Carl Zeiss EVO LS10, UK) to
Figure 2 shows the voltage outputs’ change from start to stable operation of the MFCs under different MF intensities and directions. The red arrows represent the first time for MFCs to reach the maximum power production. Cell voltages over subsequent cycles were then reproducible in terms of maximum voltage. When the MFCs were exposed to attractive MFs (Fig. 2a), the time required to reach the first maximum power cycle was significantly shorter than the control. MFC1a with an attractive MF of 50 mT started fastest; it required 10 days before reaching maximum power production. The maximum voltage under these conditions was 553±2 mV. For MFC2a and MFC3a with attractive MFs of 100 and 200 mT, respectively, the times required to reach the first maximum power cycle both increased to ∼13 days, and the maximum voltages were 536±2 and 517±1 mV, respectively. The time required for MFC0 to reach the first maximum power cycle was 16 days, which was the longest, and the maximum voltage was 523±2 mV. These results suggest that the attractive MFs accelerated the start-up. In addition, compared to the control (MFC0), a slight increase in the maximum output voltage was observed for MFC1a and MFC2a. However, the maximum output voltage for MFC3a did not increase but decreased a little. The results indicate that the attractive MF application is of benefit to electricity production, but the intensity of the applied MF should not be too high; otherwise, it would have negative effects on electricity production. To further explore the effects of MF direction on the startup and electricity production of MFCs, repellant MFs were applied to MFC1r, MFC2r, and MFC3r. Figure 2b shows the voltage outputs’ change from start to stable operation of the
9882
Appl Microbiol Biotechnol (2014) 98:9879–9887 A
B
0.6
0.5 MFC0
0.4
Voltage(V)
Voltage(V)
0.6
MFC1a
0.5 MFC0
0.3 0.2 0.1
0.5 MFC2r
MFC3a
Voltage(V)
0.4 0.3 0.2
0.3 0.2
0.0
5000 10000 15000 20000 250000
5000 10000 15000 20000 25000
Time (min)
Time (min)
MFCs under repellant MFs. No distinct difference was found between MFC1a and MFC1r, MFC2a and MFC2r, and MFC3a and MFC3r in terms of start-up time and maximum voltage. This suggests that the MF direction exhibited negligible influence on start-up and voltage output in this work. Phosphorus and nitrogen removal and crystal accumulation In this study, the effect of MF on phosphorus removal, nitrogen removal, and transformation was investigated. Concentrations of TP, N-NH4+, N-NO3−, and N-NO2− in the effluent from the anode chamber and cathode chamber of MFC0, MFC1a, MFC2a, and MFC3a were measured, and results are shown in Table 1. Nitrogen and phosphorus in the synthetic wastewater were in the form of N-NH4+ and P-PO43− in the beginning. The concentrations of N-NH4+ and P-PO43− were 55.0 and 122.9 mg/L, respectively. The concentrations of TP, NNO3−, and N-NO2− in the effluent of anode chambers were nearly the same as those of the synthetic water. This is probably due to the low DO in the anolytes, which led to oxygen shortage for bacteria in anode chambers aerobic phosphorus uptake and ammonia nitrogen nitrification or nitrosification. However, the concentration of TN in the effluent of anode Table 1 TP, N-NH4+, N-NO3−, and N-NO2− concentrations in the effluents from MFCs under different attractive MF intensities
Effluent of cathode chamber
MFC3r
0.4
0.1
0
Effluent of anode chamber
0.2
0.6
0.5 MFC2a
0.0
MFC0 MFC1a MFC2a MFC3a MFC0 MFC1a MFC2a MFC3a
0.3
0.0
0.6
0.1
Synthetic water
MFC1r
0.4
0.1
0.0
Voltage(V)
Fig. 2 MFC voltage outputs change during the start-up period. a MFC1a, MFC2a, and MFC3a under attractive MFs of 50, 100, and 200 mT, respectively; b MFC1r, MFC2r, and MFC3r under repellant MFs of 50, 100, and 200 mT, respectively. MFC0 without applying MF was run as the control throughout the study
TP (mg/L)
N-NH4+ (mg/L)
N-NO3− (mg/L)
N-NO2− (mg/L)
122.9 120.7±2.5 122.1±2.6 121.3±1.8 120.5±3.3 8.4±1.7 4.8±0.3 5.1±0.7 5.7±0.4
55.0 44.8±1.2 40.2±1.5 41.9±0.7 43.4±2.0 5.2±0.9 3.6±0.8 4.1±0.5 4.7±1.1
0 ≈0 2.6±1.0 2.2±0.8 2.5±0.7 1.3±0.5 37.2±1.6 36.1±2.3 35.8±2.4
0 ≈0 ≈0 ≈0 ≈0 34.8±2.7 0.5±0.3 0.4±0.2 0.7±0.1
0
5000 10000 15000 20000 250000 Time (min)
5000 10000 15000 20000 25000 Time (min)
chambers decreased about 10 mg/L compared with that of synthetic wastewater. The decrease of TN may be due to the consumption for microorganisms’ growth and reproduction, ammonium ion diffusion from the anode chamber to the cathode chamber, or ammonium volatilization. In comparison with MFC0, there was no significant difference in the concentrations of TP, N-NH4+, and N-NO2− in the effluent from anode chambers except a small quantity of N-NO3− presented in anode chambers of magnetic MFCs. TP removal efficiencies in MFC0, MFC1a, MFC2a, and MFC3a were ∼93, ∼96, ∼96, and ∼95 %, respectively. Compared with the control, there was a slight increase in TP removal as a MF was applied to MFCs. Some sediment accumulated on the inner walls of cathode chambers after operating for a period of time. Sediment was characterized by SEM-EDS. The results of MFC0 and MFC1a are shown in Figs. 3 and 4 (results of MFC2a and MFC3a were not shown for their similarity). SEM images show that sediment in cathode chambers was a mixture of microbe cells and chemical precipitates. The EDS patterns from the chemical precipitates in the cathode chambers of MFC0 and MFC1a are similar. The chemical precipitates were analyzed to be a mixture of phosphate, carbonate, and hydroxyl compound in an earlier study (Tao et al. 2014). Chemical precipitates formed in cathode chambers most likely resulted from the progressively alkaline catholyte. The pH values of the effluent from cathode chambers of MFC0, MFC1a, MFC2a, and MFC3a were 9.27 ±0.11, 9.38±0.13, 9.35±0.15, and 9.33±0.14, respectively. The slight increase in TP removal may be attributed to the higher pH in the cathode chamber of magnetic MFCs. The precipitation of phosphorus could be promoted in the presence of MF. TP was probably removed by microbe absorption and chemical precipitation, and the latter may be the major factor (Tao et al. 2014). Though there was almost no TN removal in all cathode chambers, the rate of N-NH4+ converted to N-NO3− or NNO2− was very different. In the control, a large proportion of N-NH4+ was converted to N-NO2−. In magnetic MFCs, however, the vast majority of N-NH4+ was converted to N-NO3−. The results show that there does exist the possibility of partial
Appl Microbiol Biotechnol (2014) 98:9879–9887
9883
Fig. 3 SEM images of sediment deposit on the inner walls of the cathode chamber of MFC0 (a) and MFC1a (b)
nitrification in the cathode chamber of MFC0 but more thorough nitrification in the cathode chambers of magnetic MFCs. Substrate degradation and coulombic efficiency COD removal efficiencies and coulombic efficiencies of magnetic MFCs and control were calculated. As shown in Fig. 5, considerably higher COD removal efficiencies were witnessed in the magnetic MFCs compared to the control (∼80 %). Moreover, the highest COD removal efficiency (>90 %) was obtained in MFC1a (under a MF of 50 mT). Despite the better degradation performance, the coulombic efficiencies of the magnetic MFCs were relatively lower than those of the control. The results indicated magnetic systems with a higher percentage of electrochemically inactive microorganisms. Polarization and power density curves The polarization curves of the magnetic MFCs and the control are illustrated in Fig. 6. The maximum power densities obtained in MFC0, MFC1a, MFC2a, and MFC3a were 526, 548, 539, and 523 mW/m2, respectively. The internal resistances were 232, 207, 221, and 241 Ω, respectively. The results imply that low-intensity MF (e.g., 50 and 100 mT) could reduce the MFC internal resistance and then improve the maximum power density. High-intensity MF may have negative effects Fig. 4 Energy-dispersive spectrographs of precipitate in SEM image a and b
on internal resistance and power density. The polarization and power density curves also demonstrate a decrease in activation losses under a MF at an appropriate intensity (50 or 100 mT in this work). It implies that the metabolic and electrochemical activities of some microorganisms were likely to be improved under MF exposure, which lowered the activation loss. This conclusion was supported by the higher COD removal in magnetic MFCs. Persistency effects of MF on MFC The persistency effects of MF on electricity generation and TP, N-NH4+, N-NO3−, N-NO2−, and COD removal were explored in this work. The changes of the output voltage of magnetic MFCs with varied MFs and control are demonstrated in Fig. 7a. MFC1a and MFC4 were operated under MFs of 50 and 0 mT at the first stage. In comparison with MFC0 and MFC4, a slight increase in the maximum output voltage was observed for MFC1a at stage 1. For MFC4, a sudden increase (though the increase was modest) of voltage was observed when a MF was applied at stage 2, but the voltage dropped sharply to less than 50 mV at stage 3 when the MF was removed. For MFC1a, the voltage suddenly dropped sharply to less than 50 mT when the MF was removed at stage 2, but the voltage rose again to the original level when resuming the MF at stage 3. The immediate increase in voltage for MFC4 after employing a MF at stage 2 and for MFC1a after
B
A O Mg
P
O
Mg
P
C Na
0
Mn
Mn
K
C Na
5
0
5
9884
Appl Microbiol Biotechnol (2014) 98:9879–9887
100
100
80 COD removal efficiency Coulombic efficiency
60
60
40
40
20
20
0
Coulombic efficiency (%)
COD removal efficiency (%)
80
(the results are in Table 1). The residual TP, N-NH4+, N-NO3−, and N-NO2− in the effluent from the cathode chamber of MFC4 and MFC1a were at the same levels as that in the control when not employing a MF to MFC1a at stage 2 and to MFC4 at stages 1 and 3. However, diversified COD removal efficiency was observed for MFC4 and MFC1a with a varied MF at three stages. Higher COD removal efficiency was obtained after employing a MF for MFC4 at stage 2 and for MFC1a at stages 1 and 3, but dropped sharply to lower than that in the control for MFC 4 at stage 3 and for MFC1a at stage 2 when the MF was removed. The above results indicate that the effect of MF on wastewater treatment in MFCs was also immediate and reversible. The higher COD removal efficiencies for MFC4 at stage 2 and for MFC1a at stages 1 and 3 show that MF plays a positive role in promoting COD removal. However, the drop of COD removal efficiencies to lower than that in the control for MFC 4 at stage 3 and for MFC1a at stage 2 indicates that there may be some negative effects when the MF was removed. Approximately 70 % COD was removed in MFC1a at stage 2 and in MFC4 at stage 3; at this moment, however, the voltage of MFC1a and MFC4 was less than 50 mV; this suggests that the great mass of COD was removed by electrochemically inactive microorganisms.
0 0
100 50 Magnetic field strength (mT)
200
Fig. 5 COD removal efficiency and coulombic efficiency of MFC under different attractive MF intensities
resuming a MF at stage 3 indicates that the enhancement effect of MF seems to be immediate and reversible. In addition, the rapid recovery of electricity production after resuming the MF suggests that the MF may promote the electron transfer or catalytic activity of electro-microorganisms rather than causing irreversible changes to the cell structure. The sharp decline of voltage after removing the MF for MFC4 at stage 3 and for MFC1a at stage 2 implies that the positive effect of MF on electricity production seemed to be nonpersistent and even has negative effects when suddenly pausing the MF, possibly leading to a period of shutdown for the microorganisms that have adapted to the MF condition. The sharp decline of voltage after MF removal should be mainly caused by a pause of MF rather than the substrate depletion. This was supported by the relatively high voltage in the control. In order to validate the long-term effect of MF on wastewater treatment, the residual TP, N-NH4+, N-NO3−, and NNO2− and COD removal efficiency in the effluent from cathode chambers were investigated. The results are shown in Fig. 7b. As seen from the results in Fig. 7b, when a MF was applied to MFC4 at stage 2 and to MFC1a at stages 1 and 3, the residual TP, N-NH4+, N-NO3−, and N-NO2− in the effluent from cathode chambers of MFC4 and MFC1a were similar to those of MFC1a with a constant MF of 50 mT in the first test
The magnetic MFCs started faster than the control under MF exposure, with a difference that varied within 10–13 days, depending on the MF intensity applied (Fig. 2). Moreover, enhancement of electricity production was observed for MFC1a and MFC2a but not for MFC3a. All of these results demonstrated that MF application could accelerate start-up and promote electricity production of MFCs. It was probably because the MFs promoted the enrichment of anode electrogenesis and accelerated the rate of proliferation. The applied MF may reduce anode activation loss, which results from microbial metabolic activity and electrochemical reactions
A
B 700
600
MFC0 MFC1a MFC2a MFC3a
MFC0 MFC1a MFC2a MFC3a
500
Power density (mW/m )
600
2
500
Voltage (mV)
Fig. 6 Polarization (a) and power density (b) curves of MFC under different attractive MF intensities
Discussion
400 300 200 100
400
300
200
100
0
500
1000
1500
2000
2500 2
Current density (mA/m )
3000
3500
0
500
1000
1500
2000
2500 2
Current density (mA/m )
3000
3500
Appl Microbiol Biotechnol (2014) 98:9879–9887
A
Stage1 0.6
Voltage (V)
Fig. 7 Variations of output voltage (a), residual TP, N-NH4+, N-NO3−, N-NO2−, and COD removal efficiency in the effluent from cathode chambers (b) of magnetic MFCs with varied MFs and control
9885
Stage3 without a MF
MFC0
0.2 0.0 without a MF
apply a MF
remove the MF
MFC4
0.4 0.2 0.0 apply a MF
0.6
Voltage (V)
Stage2 without a MF
0.4
0.6
Voltage (V)
without a MF
remove the MF
re-apply a MF
MFC1a 0.4 0.2 0.0 0
2500
5000
7500
10000
12500
15000
17500
20000
Time (min)
concentration (mg/L)
+ N-NH4
TP
Stage3
N-NO2
COD
80 60 40 20 0 80 60 40 20 0
COD removal efficiency (%)
40 30 MFC0 20 10 0 40 30 MFC4 20 10 0 40 30 MFC1a 20 10 0 2500
N-NO3
COD removal efficiency (%)
concentration (mg/L)
Stage2
Stage1
COD removal efficiency (%)
concentration (mg/L)
B
80 60 40 20 0 5000
7500
10000
12500
15000
17500
20000
Time (min)
(Yin et al. 2013). However, the intensity of the applied MF should not be too high, or it may have less effect and even negative effects on start-up and electricity production. Nitrogen compound transformations in magnetic MFCs and control were very different. Results in Table 1 show that the vast majority of residual TN in the effluent from cathode chambers of magnetic MFCs was N-NO3−, but in the control, it was N-NO2−. Nitrification is a two-step process: (1) NH4+ or
ammonia (NH3) is oxidized into NO2− by ammonia-oxidizing bacteria (AOB), often Nitrosomonas spp, and (2) the NO2− is further oxidized into NO3− by nitrite-oxidizing bacteria (NOB), often Nitrobacter spp. The nitrification process was stopped at the end of the first step maybe because of the shortage of oxygen in the control. Nitrification was carried out in the magnetic MFCs in a more complete way. This was consistent with previous research of Tomska and Wolny
9886
(2008), who found that more intensive nitrogen compound transformations and higher nitrification rate could be obtained in magnetic-activated sludge units. In addition, most of this was due to the higher oxygen uptake rate in the second nitrification phase. Many studies have suggested that MF is an intensifying factor for organic substrate degradation (Jung and Sofer 1997; Jung et al. 1993). The study of Li et al. (2011) has also shown that the MF could stimulate cell growth. One potential explanation for such improvement of microbial activity by MF is that the MF causes an oxidative stress and increases the activity, lifetime, and concentration of free radicals (Zhang et al. 2003; Jones et al. 2007). In consequence, highly reactive byproducts of usual metabolism and immune defense such as oxidative metabolites, free radicals, and reactive oxygen species (Zhang et al. 2003) increase and cause a change in enzymatic activity. Therefore, the possible mechanism could be involved with the change of bioelectrochemical reaction rate upon MF application to MFCs. A further exploration of the persistency effects of MF was carried out by applying a varied MF to the MFCs. Results in previous sections suggested that the MF effects on MFC performance for electricity generation and wastewater treatment were immediate, reversible, and not long lasting. Negative effects on electricity generation and COD removal would occur when the MF is suddenly paused. That may be because the biological behaviors of some microorganisms in magnetic MFCs have been altered; then, they will not adapt to the living environment without employing a MF. There was no adverse effect on TP removal, nitrogen removal, and transformation but on electricity generation. That means that microorganisms that take part in TP removal, nitrogen removal, and transformation in MFCs contributed little to electricity generation in this work. This conclusion was consistent with the recent research of ours. Previous research has shown that there was no nitrifying bacteria or anaerobic ammonia oxidation bacteria in the anode chamber for the long time anaerobic environment in the anode chamber. When the cathodic DO was at 3.5 mg/L, N-NO3− and N-NO2− in the catholyte would not be used as the electron acceptors for electricity generation (Tao et al. 2014). Acknowledgments Financial supports from the National Natural Science Foundation (21277052), the Department of Guangdong Education and the Science and Technology Bureau (2010), the State Key Laboratory of Subtropical Building Science (2013ZC03, 2014ZB04), and the Environmental Protection Bureau (201203) are gratefully acknowledged.
References APHA (2005) Standard methods for the examination of water and wastewater, 19th edn. American Public Health Association, Washington
Appl Microbiol Biotechnol (2014) 98:9879–9887 Butler CS, Nerenberg R (2010) Performance and microbial ecology of air-cathode microbial fuel cells with layered electrode assemblies. Appl Microbiol Biotechnol 86:1399–1408 Chen SS, Liu GL, Zhang RD, Qin BY, Luo Y, Hou YP (2012) Improved performance of the microbial electrolysis desalination and chemicalproduction cell using the stack structure. Bioresour Technol 116: 507–511 Cusick RD, Logan BE (2012) Phosphate recovery as struvite within a single chamber microbial electrolysis cell. Bioresour Technol 107: 110–115 Domagalski J, Lin C, Luo Y, Kang J, Wang S, Brown LR, Munn MD (2007) Eutrophication study at the Panjiakou-Daheiting reservoir system, northern Hebei province, People’s Republic of China: chlorophyll-A model and sources of phosphorus and nitrogen. Agr Water Manag 94(1–3):43–53 Ichihashi O, Hirooka K (2012) Removal and recovery of phosphorus as struvite from swine wastewater using microbial fuel cell. Bioresour Technol 114:303–307 Ichihashi O, Yamamoto N, Hirooka K (2012) Power generation by and microbial community structure in microbial fuel cell treating animal wastewater. J Jpn Soc Water Environ 35(1):19–26 Jiang D, Li B, Jia W, Lei Y (2010a) Effect of inoculum types on bacterial adhesion and power production in microbial fuel cells. Appl Biochem Biotechnol 160(1):182–196 Jiang D, Li X, Raymond D, Mooradain J, Li B (2010b) Power recovery with multi-anode/cathode microbial fuel cells suitable for future large-scale applications. Int J Hydrog Energy 35(16): 8683–8689 Jiang D, Curtis M, Troop E, Scheible K, McGrath J, Boxun H, Suib S, Raymond D, Baikun L (2011) A pilot-scale study on utilizing multianode/cathode microbial fuel cells (MAC MFCs) to enhance the power production in wastewater treatment. Int J Hydrog Energy 36(1):876–884 Jones AR, Hay S, Woodward JR, Scrutton NS (2007) Magnetic field effect studies indicate reduced geminate recombination of the radical pair in substrate-bound adenosylcobalamin-dependent ethanolamine ammonia lyase. J Am Chem Soc 129(50):15718–15727 Jung JT, Sofer S (1997) Enhancement of phenol biodegradation by south magnetic field exposure. J Chem Technol Biotechnol 70(3):229– 303 Jung JT, Sanji B, Godbole S, Sofer S (1993) Biodegradation of phenol: a comparative study with and without applying magnetic fields. J Chem Technol Biotechnol 56(1):73–76 Li WW, Sheng GP, Liu XW, Cai PJ, Sun M, Xiao X, Wang YK, Tong ZH, Dong F, Yu HQ (2011) Impact of a static magnetic field on the electricity production of Shewanella-inoculated microbial fuel cells. Biosens Bioelectron 26(10):3987–3992 Liu H, Logan BE (2004) Electricity generation using an air-cathode single chamber microbial fuel cell in the presence and absence of a proton exchange membrane. Environ Sci Technol 38(14):4040– 4046 Liu H, Cheng SA, Logan BE (2005) Power generation in fed-batch microbial fuel cells as a function of ionic strength, temperature, and reactor configuration. Environ Sci Technol 39(14): 5488–5493 Liu ST, Yang FL, Meng FG, Chen HH, Gong Z (2008) Enhanced anammox consortium activity for nitrogen removal: impacts of static magnetic field. J Biotechnol 138(3–4):96–102 Logan BE (2009) Exoelectrogenic bacteria that power microbial fuel cells. Nat Rev Microbiol 7(5):375–381 Lovley DR (2008) The microbe electric: conversion of organic matter to electricity. Curr Opin Biotechnol 19(6):564–571 Nelson NO, Mikkelsen RL, Hesterberg DL (2003) Struvite precipitation in anaerobic swine lagoon liquid: effect of pH and Mg: P ratio and determination of rate constant. Bioresour Technol 89(3):229–236
Appl Microbiol Biotechnol (2014) 98:9879–9887 Puig S, Serra M, Vilar-Sanz A, Cabré M, Bañeras L, Colprim J, Balaguer MD (2011) Autotrophic nitrite removal in the cathode of microbial fuel cells. Bioresour Technol 102(6):4462–4467 Puig S, Coma M, Desloover J, Boon N, Colprim J, Balaguer MD (2012) Autotrophic denitrification in microbial fuel cells treating low ionic strength waters. Environ Sci Technol 46:2309–2315 Rabaey K, Lissens G, Siciliano SD, Verstraete W (2003) A microbial fuel cell capable of converting glucose to electricity at high rate and efficiency. Biotechnol Lett 25(18):1531–1535 Rozendal RA, Leone E, Keller J, Rabaey K (2009) Efficient hydrogen peroxide generation from organic matter in a bioelectrochemical system. Electrochem Commun 11(9):1752–1755 Tao QQ, Luo JJ, Zhou J, Zhou SQ, Liu GL, Zhang RD (2014) Effect of dissolved oxygen on nitrogen and phosphorus removal and electricity production in microbial fuel cell. Bioresour Technol 164:402–407 Tomska A, Wolny L (2008) Enhancement of biological wastewater treatment by magnetic field exposure. Desalination 222(1–3): 368–373 Virdis B, Rabaey K, Yuan Z, Keller J (2008) Microbial fuel cells for simultaneous carbon and nitrogen removal. Water Res 42(12):3013– 3024
9887 Virdis B, Rabaey K, Yuan Z, Rozendal RA, Keller J (2009) Electron fluxes in a microbial fuel cell performing carbon and nitrogen removal. Environ Sci Technol 43(13):5144–5149 Virdis B, Rabaey K, Rozendal RA, Yuan Z, Keller J (2010) Simultaneous nitrification, denitrification and carbon removal in microbial fuel cells. Water Res 44(9):2970–2980 Watanabe K (2008) Recent developments in microbial fuel cell technologies for sustainable bioenergy. J Biosci Bioeng 106(6): 528–536 Yavus H, Celebi SS (2000) Effects of magnetic field on activity of activated sludge in wastewater treatment. Enzym Microbiol Technol 26(1):22–27 Yin Y, Huang GT, Tong YR, Liu YD, Zhang LH (2013) Electricity production and electrochemical impedance modeling of microbial fuel cells under static magnetic field. J Power Sources 237: 58–63 Zhang QM, Tokiwa M, Doi T, Nakahara T, Chang P, Nakamura N, Hori M, Miyakoshi J, Yonei S (2003) Strong static magnetic field and the induction of mutations through elevated production of reactive oxygen species in Escherichia coli soxR. Int J Radiat Biol 79(4): 281–286
