The Electricity Generation in Microbial Fuel Cells Using Reaeration Mechanism for Cathodic Oxygen Reduction

Electricity generation in microbial fuel cell (MFC) using reaeration mechanism to facilitate cathodic oxygen reduction is sustainable and economical. This study examined the effects of operational parameters of electrical load (Rext), organic load and cathode area (Sa) on MFC performance under reaeration rate (K2) of 0.5-1.5 d -1in cathode chamber. Two MFCs, consisting of MFC-A (with Rext10 Ω) and MFC-B (Rext1000 Ω), were operated in parallel and continuously fed with influent chemical oxygen demand (CODin) 324–561 mg/L to anode chamber; and in each MFC the Sa covering 184, 553, 992 and 1290 cm2 was tested. Results indicated that in MFC-A the current production increased with aqueous COD in anode chamber, in which the relationship between current and aqueous COD can be modeled with Monod kinetics. The estimated kinetic constants of maximum current Imax is 3 mA, and half-saturation constant of current Ks is 310 mg/L. The lowest dissolved oxygen (DO) of 1.9 mg/L occurred at highest CODin of 561 mg/L. In MFC-B, constant current of 0.4 mA and DO at 3.2-3.7 mg/L were maintained for all CODin. The Sa had insignificant influence on electricity generation in both MFCs. This study demonstrated the importance of electrical load, organic load, and their interactions among them in designing reaeration-assisted MFC for organic waste treatment. *Corresponding author: Chi-Yuan Lee, Water Resources and Environmental Engineering Program, Department of Harbor and River Engineering, National Taiwan Ocean University, Keelung 20224, Taiwan, Tel: +886-2-2462-2192 ext. 6147; E-mail: cylee@mail.ntou.edu.tw Received November 20, 2015; Accepted December 04, 2015; Published December 14, 2015 Citation: Lee CY, Lin YH (2015) The Electricity Generation in Microbial Fuel Cells Using Reaeration Mechanism for Cathodic Oxygen Reduction. J Civil Environ Eng 5: 203. doi:10.4172/2165-784X.1000203 Copyright: © 2015 Lee CY, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


Introduction
Reaeration is a natural process in which oxygen is transferred from the atmosphere to water body when dissolved oxygen (DO) is under saturated. The use of reaeration process in rivers for partially treating and disposing of waste has been practiced for several decades. A key issue of avoiding DO drop to critical condition during discharging waste to rivers was to calculate the maximal organic loads related to reaeration coefficient, K 2 . The K 2 is dependent on stream characteristics including the flow velocity, water depth, and channel slope, varied substantially between 0.1 d -1 and 50 d -1 [1][2][3].
A microbial fuel cell is a promising system that directly converts chemical energy in organic substrate into electricity, with advantages in recovering electrical energy during wastewater treatment. The most challenge in commercializing MFCs is to develop sustainable processes that are cost effective. In a typical two-chambered MFC system mechanical aeration is usually used for supplying dissolved oxygen in cathode chamber, which is energy intensive. However, if a MFC is built on river bank or costal area, the water flow can be diverted into cathode chamber, thus, reaeration mechanism [4] can be applied for supplying oxygen to facilitate cathodic reaction. The novel reaeration-assisted MFC had latent benefit of aeration energy savings, where about 1 kg O 2 / m 3 transferred is equivalent to the production of 1 kWh/m 3 [5]. In addition, the proposed MFC only abstracts the oxygen in river, but avoiding the direct contact of pollutants with water, thus the water quality can be better protected. Though numerous studies on the relationship of MFC performance to dissolved oxygen by mechanical aeration were conducted previously [6,7], no investigation has ever been done on revealing the performance of MFC related to reaeration rate. Except for reaeration rate, in practical application of MFC to wastewater treatment the electricity generation can be affected by many factors, such as organic loads, electrical loads (external resistance, R ext ), and cathode area [8][9][10][11][12][13]. In this study, under a basic reaeration rate of still water surface, two MFCs with different R ext were employed and operated in parallel for the purpose of examining MFC performance in response to the changes in influent chemical oxygen demand (COD in ) and cathode area (S a ). The specific purpose of this study is three-fold: first to compare the effects of these operational parameters on the electricity generations; second to examine the importance of electrical load in the design of reaeration-assisted MFC; and third to verify the prediction of current generation based on oxygen supply rate.

Current generation in reaeration-assisted MFC
The current generation in reaeration-assisted MFC related to oxygen supply rate in cathode chamber can be analyzed using the mass balance of DO in cathode chamber, which is expressed as Where K 2 is the oxygen reaeration coefficient (d -1 ), DO* is the saturated DO (e.g., 8.2 mg/L at 25°C), and OR is the cathodic reduction rate (oxygen consumption rate) during electricity generation. Suppose that the changes in DO are in a steady state in cathode chamber, dDO/ dt = 0, yielding OR = K 2 (DO*-DO). (2) The above equation indicates that in cathode chamber at steady state, oxygen consumption rate for producing current, OR, equals the oxygen transfer rate from reaeration mechanism, K 2 (DO*-DO). That is, based on oxygen half-reaction, one mole of O 2 transferred will produce 4 eq e -: Equation (1-3) is a theoretical basis that can be used to predict the relationship of current production to oxygen transfer rate in reaeration-assisted MFC.

MFC construction
Two MFC systems consisting of MFC-A and MFC-B were employed in this study, where MFC-A was loaded with R ext of 10 Ω and MFC-B with1000 Ω. The MFC was constructed of acrylic with a volume of 0.6 L in an anode chamber (14 cm in length × 10 cm in width × 8.5 cm in depth) and of 1.7 L in a cathode chamber (14 cm × 12.6 cm × 9.5 cm, respectively). The anodic and cathodic chambers were separated using a 7 cm × 7 cm Nafion membrane (NRE212, DuPont, USA), and a copper wire loaded with an external resistance (electrical load) was used to connect the anode and cathode. The total graphite surface area in the anodic chamber was 790 cm 2 , consisting of one plain graphite plate of 184 cm 2 and graphite granules of 606 cm 2 . In the cathodic a specific number of chamber plain graphite plates were placed for each test run, including 1 (surface area, S a, 184 cm 2 ), 3 (S a, 553 cm 2 ), 5 (S a, 992cm 2 ), and 7 (S a, 1290 cm 2 ). The schematic diagram of experimental set-up is shown in Figure 1.

MFC operation
Before the MFC was operated, the anode was seeded with bacteria in soil, extracted 30 cm from the surface in a public park in Keelung, Taiwan. The soil samples were pretreated by performing heat-shock at 104°C in an oven for 2 h. The samples were then sieved through a #20 mesh, and stored in bottles in a refrigerator (4°C) until use [14]. During bacteria seeding, the anode was inoculated with electricigens in the soil samples with the addition of 5 g to the anode chamber. Artificial organic waste was prepared by dissolving sodium acetate and inorganic nutrient into tap water pretreated by conducting dechlorination. The nutrient and buffer components for the fuel consisted of 13 mg/L of NaH 2 PO 4 ‧H 2 O, 48 mg/L of NH 4 Cl, 7.6 mM of KH 2 PO 4 , and 42.4 mM of Na 2 HPO 4 . In the experiment, the influent concentrations of 324-561 mg COD/L was continually discharged into the anode chamber at a flow rate of 0.6 L/d. The hydraulic retention times was 1 day based on the volume of anode chamber, equivalent to organic loading rate of 0.3-0.6 kg COD/m 3 cathode chamber-d. The electrolyte in the cathode chamber consisted of 50 mM of H 3 PO 4 and 100 mM of NaCl.

Analysis
The reaeration coefficient (K 2 ) was determined using absorption measurement, where DO probe (YSI) was installed 15 cm from the bottom of cathode chamber [15]. Influent and effluent COD were monitored periodically using the Hach technique (Hach Company, USA). Voltage was measured using a CHY model-48R digital multimeter (CHY Firemate Co., Ltd., Taiwan). Current and power were calculated according to the following equations: I=V/R ext and P=IV, respectively, where I is current (A), V is the voltage (V), R ext is the electrical load (Ω), and P is power (W). We placed an Ag/AgCl reference electrode 0.5-1 cm from the cathode to measure the cathode potential. The polarization test was conducted by placing a series of resistors ranging from 1 to 3000 kΩ as the external resistance. Internal resistance was calculated using R int = (OCV-E cell )/I L , where OCV is open-circuit voltage, E cell is the cell voltage, and I L is the current. Coulombic efficiency (CE%) was derived as follows: CE(%) = the measured coulombs/theoretical coulombs of COD removals.

Determination of K 2
All of the test runs for evaluating MFC performance were conducted under the reaeration conditions of still water surface and at room temperature (24-26°C). The electrolyte in the cathode chamber at a still water surface represents a basic condition, enabling us to evaluate the critical generation of electricity. Figure 2 shows the oxygen deficit as a function of elapse time in a typical reaeration coefficient test at 25°C. Based on this oxygen deficit curve, the reaeration coefficient (K 2 ) was determined using linear regression (R 2 =0.98) to be 0.8 d -1 (0.034 h -1 ). All of the tested K 2 values in this study were in the range of 0.5-1.5 d -1 , which is similar to other report [16].

Performance comparisons
The two MFCs were discharged with influent concentration of 324-561 mg COD/L, equivalent to organic loadings of 0.3-0.6 kg COD/m 3 cathode chambered. These loading rates, within the levels of conventional activated sludge process and anaerobic contact process, were selected to insure that the electron flow from anode to cathode was not a limiting factor in cathodic oxygen reduction. It was observed that in MFC-A the aqueous COD in anode chamber (COD eff ) was maintained at 232-355 mg/L, whereas in MFC-B it was at 247-447 mg/L. The two MFCs had similar COD removal efficiencies, 23-44% v.s. 17-41%, implied that the electrical loads was indifferent to organic removal. To improve COD removal, it can be done by decreasing organic load but not changing electrical load. The voltage generations in the two MFCs were different, as compared in  (Table 1). These results clearly indicated that electrical load had significant impact on the electricity generations.

Effects of organic load
The effect of organic load for each MFC was tested by changing the influent COD. In MFC-A, the current generation was 1.0 mA at CO-D in 324 mg/L, and the current became doubled when COD in increased to 516 mg/L. Since aqueous COD in the anode chamber determines the substrate flux diffused to anode biofilm, it is practical to use CO-D eff instead of COD in as a key parameter in analyzing the relationship of substrate concentration to current generation [17]. Figure 4 shows that current generation strongly depends on substrate concentration in terms of COD eff , and a Monod kinetics modeling could be well established. The estimated kinetic constants of maximum current I max is 3 mA, and half-saturation constant of current K s is 310 mg/L. This result clearly showed that under low electrical load, applied COD was a limiting factor in controlling current generation. Furthermore, it was observed that the DO decreased from 2.8 mg/L to 1.9 mg/L (cathode potential varied from -11 to -216 mV) when COD in increased from 324 to 516 mg/L. The DO level appeared to be negatively correlated with current generation, where the R 2 for current against DO regression was 0.64 ( Figure 5). The low DO yielding high current production in reaeration assisted-MFC could be attributed to the fact that low DO gives high oxygen transfer rate (Equations 1-3). Notably, the lowest DO of 1.9 mg/L occurred at COD in 516 mg/L, implying that subjected to the highest organic load, the cathode chamber was exposed to the highest risk of anaerobic condition. A quite different outcome occurred in MFC-B, where current generation was maintained at the level of 0.4 mA, regardless of variations in COD in , implying that under high electrical load the influent COD lose its influence on electricity generation. With high electrical load in MFC-B the electron transfer from anode to cathode was limited by the electrical resistance, resulting in insignificant variations of current production. Thus, the DO could be maintained at a relatively high level, 3.2-3.7 mg/L (cathode potential varied from +64 to +170 mV), a condition can be assured completely aerobic.

Effects of cathode area
The limitation of cathodic reaction was tested with increasing cathode surface. It was found that in MFC-A the current was 0.9 mA at S a 184 cm 2 and slightly increased to 1.0 mA when S a was greatly increased to 1290 cm 2 . This result indicated that S a had negligible effect on current generation. Similarly, it showed that cathode surface area was not a limiting factor in current generation for MFC-B, where the current was maintained at 0.4 mA regardless of S a increasing from 184 cm 2 to 1290 cm 2 . The above results demonstrated that under the reaction rate of 0.5-1.5, plain graphite with S a 184 cm 2 was sufficient for cathodic reduction. Though the plain graphite had been criticized for being poor in catalyzing oxygen reduction, it is demonstrated herein that this material is suitable to facilitate cathodic reaeration when low reaeration coefficient of 0.5-1.5 d -1 was employed as passive oxygen supply.

Correlation of current generation with oxygen transfer
The extreme currents can be generated through reaeration mechanism is computed using Equations 1-3. Supposed that the reaeration rate is 0.5 d -1 , the maximum oxygen can be supplied in cathode chamber is 0.5(8. . This current represents theoretical production in reaeration-assisted MFC operated at K 2 0.5 d -1 . In case the reaeration rate increased to 1.5 d -1 , the maximum oxygen supply rate is 0.66 mM O 2 /d and equivalent electron flow of 2.64 meq e -/d (3.0 mA). Obviously, the current at K 2 of 1.5 d -1 increased 3 times than that at the reaeration rate of 0.5 d -1 . The predicted current 3.0 mA coincides with I max , which confirms the maximum current generation in reaeration-assisted MFC at still water surface would be 3.0 mA. Table 2 further lists the I eqv values computed from oxygen supply in response to K 2 and DO. Again, the I eqv appeared to increase with K 2 value at a specific DO level. For example, at DO 2.8 mg/L (similar to Test run A1), the I eqv values were 0.6, 1.2, and 1.8 mA when K 2 was 0.5, 1.0, and 1.5 d -1 , respectively. In comparison of the I eqv values with the measured currents I avg (Table 1), it was apparent that the I eqv for MFC-A was within the range of the I avg , implying that the current production correlated well with oxygen transfer rate. Nevertheless, in MFC-B the I eqv was slightly higher than the I avg indicating that a few oxygen generated from reaeration mechanism was lost, which might be diffused from cathode to anode because DO in cathode chamber was at relatively high level.

Discussion
In conventional practice of discharging waste to river that has a specific reaeration rate, the applied organic load solely control DO level. However, in this study, we have demonstrated that in reaerationassisted MFC the DO not only was affected by organic loading but also by electrical load. Compared to MFC-A having low electrical load, the MFC-B with high electrical load could maintain the DO above 3.2 mg/L because high electrical load restricts the electron flow from anode to cathode. In addition, the electrical load also affected electricity     (2) and (3).The dissolved oxygen in cathode chamber (DO) is shown in Table 1; saturated DO is assumed to be 8.2 mg/L; the K 2 estimated is based on the experimental results conducted at still water surface, ranging over 0.5-1.5 d -1 ; and the volume of cathode is 1.7 L. generation. It was observed that MFC-B produced higher power output than MFC-A. This result might be attributed to the MFC-B having close proximity of electrical load to internal resistance [8]. Thus, selection an optimal electrical load in accordance with internal resistance is important in designing reaeration assisted MFC. Another important issue needed to be addressed is the impact of reaeration rate on MFC performance. Subsequent study is suggested to get deeper understanding into the effects of increasing reaeration rate on improving the performance of reaeration assisted-MFC.

Conclusions
1. The reaeration-assisted microbial fuel cell cells (MFC) that operated under reaeration rate of 0.5-1.5 d -1 are feasible for electricity generation.

3.
The current production can be estimated by oxygen transfer rate, based on reaeration coefficient and dissolved concentration in cathode chamber.

4.
The organic load, electrical load, and their interactions among them were crucial in the designing reaeration-assisted MFC. The optimal organic and electrical loads should be cautiously determined to maintain appropriate DO and improve electricity generation.