Dynamically Adaptive Control System for ... - ACS Publications

Apr 17, 2013 - By definition, this control system creates an independent electrical environment for all cells connected serially in a stack, effective...
0 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/est

Dynamically Adaptive Control System for Bioanodes in Serially Stacked Bioelectrochemical Systems Stephen J. Andersen,†,‡,∥ Ilje Pikaar,‡ Stefano Freguia,‡ Brian C. Lovell,§ Korneel Rabaey,*,‡,∥,# and René A. Rozendal‡,⊥,# †

Bilexys Pty Ltd., c/o UniQuest, Level 7, General Purpose South Building, Staff House Road, Queensland 4072, Australia Advanced Water Management Centre (AWMC), and §School of Information Technology and Electrical Engineering (ITEE), The University of Queensland, Queensland 4072, Australia ∥ Laboratory of Microbial Ecology and Technology (LabMET), Ghent University, Coupure Links 653, 9000 Ghent, Belgium ⊥ Paques BV, Post Office Box 52, 8560 AB, Balk, The Netherlands ‡

ABSTRACT: Microbial bioelectrochemical systems (BESs) use microorganisms as catalysts for electrode reactions. They have emerging applications in bioenergy, bioproduction, and bioremediation. BESs can be scaled up as a linked series of units or cells; however, this may lead to so-called cell reversal. Here, we demonstrate a cell balance system (CBS) that controls individual BES cells connected electrically in series by dynamically adapting the applied potential in the kilohertz frequency range relative to the performance of the bioanode. The CBS maintains the cell voltage of individual BES cells at or below a maximum set point by bypassing a portion of applied current with a high-frequency metal oxide semiconductor fieldeffect transistor switch control system. We demonstrate (i) multiple serially connected BES cells started simultaneously and rapidly from a single power source, as the CBS imparts no current limitation, (ii) continuous, stable, and independent performance of each stacked BES cell, and (iii) stable BES cell and stack performance under excessive applied currents. This control system has applications for not only serially stacked BESs in scaled-up stacks but also rapidly starting individual- and/or lab-scale BESs.

■

INTRODUCTION Bioelectrochemical systems (BESs) are regarded as a promising technology for electricity generation, bioremediation, and bioproduction.1−5 BESs use electrochemically active biocatalysts or microorganisms to generate electrons and drive electrochemical processes, including electrical power production, desalination, and chemical production (e.g., caustic soda). Organic molecules are commonly used as an electron source, thereby presenting an opportunity for these systems to generate value while simultaneously treating side streams and wastewater.6−9 To date, most BES studies have been performed at laboratory scale using single-cell (single anode, single cathode) devices, in both power output (“microbial fuel cell, MFC”) and power input (“microbial electrolysis cell, MEC”) modes. BESs can be operated in many configurations but, in all cases, require an anode and a cathode.4 In the example of a MEC for caustic soda production, microbial respiration is the oxidation reaction at the anode and electrochemical hydrolysis occurs at the cathode to generate a high pH catholyte. An organic-rich stream is delivered to a bioanode chamber as a feed source for the bioelectrogenic oxidation process that, in turn, drives the cathodic caustic production. Although the concept of © 2013 American Chemical Society

combining wastewater treatment and sustainable chemical production is intriguing, the technology is still in its infancy, with most laboratory and applied systems operating at the liter scale. For example, in a MEC reactor operating on site at a brewery with real wastewater substrate, Rabaey et al. reported a maximum generated current of 0.38 A at up to 704 L/day wastewater influent, producing 0.7 L/day low-strength caustic solution.6 Bioelectrochemical technology is currently in transition from milliliter- and liter-scale devices to systems capable of processing orders of magnitude more influent into useful quantities of power, chemicals, etc. To facilitate this transition, more studies are emerging to deal with BES performance and electrical connectivity in stacks and arrays.10−17 Serially connected stacks are operated at high voltage and lower current, whereas stacks that are connected in parallel are operated at high current and low voltage. Lower operating Received: Revised: Accepted: Published: 5488

January 23, 2013 April 5, 2013 April 17, 2013 April 17, 2013 dx.doi.org/10.1021/es400239k | Environ. Sci. Technol. 2013, 47, 5488−5494

Environmental Science & Technology

Article

and hydroxide ions. Cation flux (principally Na+) from the anode to the cathode provides the balance of charge, thus driving the formation of an alkaline solution (predominantly NaOH) in the cathode, as previously described.6 Cell performance was compared simultaneously in triplicate to three identical two-chamber electrochemical cells consisting of two Perspex frames with internal dimensions of 20 × 5 × 0.9 cm. The identical chambers are separated by a cation-exchange membrane (Ultrex CM17000, Membranes International, Inc., Ringwood, NJ). Graphite felt (100 cm2) was used in each anode, with stainless-steel mesh current collectors (6 mm mesh size). The cathode material consisted of fine stainless-steel cloth with stainless-steel mesh current collectors. Power was supplied with a VSP Modular 5 Channel Potentiostat (BioLogic Science Instruments, France). An Ag/AgCl reference electrode [+0.205 V versus standard hydrogen electrode (SHE); RE-5B, Bioanalytical Systems, Inc., West Lafayette, IN] was placed in each anode. Anode potentials were monitored using a National Instruments data acquisition unit. All electrode potentials are reported in volts versus SHE. CBS. The CBS was produced in consultation with Tritium Pty Ltd. (Tennyson, Queensland, Australia). By definition, this control system creates an independent electrical environment for all cells connected serially in a stack, effectively allowing for each cell to be assessed as an individual unit independent of other connected cells. The CBS uses two STD150N3 Nchannel power MOSFETs (STMicroelectronics, France) in cascode configuration. These surface mount transistors are characterized by high current (80 A) and extremely low resistance (0.0024 Ω). This transistor pair allows the CBS to either connect the cell to the current source or bypass the cell and pass the current directly to the next node in the string. The CBS hardware and controlling software were adapted for this specific application by modification of the IQ Battery Management System (Tritium Pty Ltd., Tennyson, Queensland, Australia).26 The hardware consists of three cell balancing nodes and an electronic communication device all controlled by personal computer (PC)-based software. User input sets the maximum cell voltage across individual BES cells. This set point remained unchanged at 1000 mV across all experiments. The three individual BES cells with CBS nodes were connected in series to a single power supply (see Figure 1), which provided a constant current through the BES stack. The CBS control system toggles the MOSFET switches at high frequencies (kilohertz range) between the open or closed states. If an individual BES cell in the stack operates at a cell voltage less than the set point of 1.0 V, the MOSFETs direct current into this BES cell. Conversely, if the individual BES cell cannot support the supplied current and the cell voltage reaches the 1.0 V set point, the MOSFETs start switching at high frequency. By doing so, the MOSFETs divert some portion of the current into the BES and the remainder through a bypass circuit to the next cell in the stack with very high power efficiency and negligible losses. In this manner, the cell voltage is set as long as there exists sufficient applied or generated current to maintain it. Further, as microbially generated current develops, the percentage of current bypassing an individual BES cell decreases as electrons are delivered to the cathode to drive the reduction reaction.

current translates to a decreased energy loss across resistances and thinner, cheaper electrical wiring. In abiotic electrochemical systems, control electronics are commonly applied to stacked devices to avoid operational problems from the inconsistent performance of single cells.18 Similarly, in BESs, inherent inequalities between cells commonly arise because of the biological nature of the electrocatalysts. In serially connected stacks, high-performing (i.e., good electrogenic activity) bioanodes can push less-performing bioanodes in the stack outside their comfort zone, which can result in unfavorable potentials during startup and continuous operation. 19,20 Further, one can expect that BES stacks and, indeed, individual bioanodes will be pushed for higher performance by increasing the applied current in the drive for higher current densities and increased output. Without effective power management, excessive bioanode potentials can lead to permanent biofilm and material damage, resulting in decreased stack performance and possibly destruction of the device. Studies with MFCs have confirmed that serial arrangements are problematic because of the unequal performance of stacked cells.21−24 For MFCs, some elegant solutions have been proposed for stacked biopower generation. For example, Kim et al. recently described a parallel cell connection using capacitors that enables the output voltage of a MFC system to be ramped up.11 However, such systems do not yet address the issue of starting up multiple BESs, in which one needs rapid development of the biocatalyst in non-limiting conditions. Here, we demonstrate a so-called cell balance system (CBS) to stabilize potentials across serially stacked MECs and prevent bioanode damage by effectively allowing for independent performance of individual BES cells, despite being connected electrically in series. Uniquely, the control system reported in this study manipulates the power input of each cell relative to the electrogenic performance of the bioanode with high responsiveness. While a number of studies concerning MFC control systems exist, to the authors’ knowledge no suitable control system for starting and operating a MEC serial stack has been developed until now.10−17 This CBS applies metal oxide semi-conductor field-effect transistor (MOSFET) switches as the central mechanism to dynamically adapt to the performance variation of bioanodes. MOSFETs can be used during charging in battery stacks to prevent overheating, overcharging, and cell degradation. MOSFETs switch at frequencies in the kilohertz range to bypass some portion of the inflowing current of each cell relative to cell performance, thereby allowing individual cells in a stack to operate at a set voltage while drip-feeding power to the BES.25 In the CBS, this mechanism is applied to balance the applied cell voltage of units within the serially connected MEC stack. Although MOSFET switches have previously been applied in a MFC power management system,12 the system did not apply the switching mechanism to balance stacked BESs relative to individual cell performance, which is critical toward the rapid startup and sustained performance of stacked BESs. This CBS is a unique and innovative approach to the practical issue of starting and operating multiple units from a single power supply and represents a solid step toward implementation of industrialscale BESs.

■

■

MATERIALS AND METHODS Bioelectrochemical Cell. In all experiments, microbial oxidation of acetate to CO2 occurred at the anode, while the cathode reaction was the reduction of water to hydrogen gas

EXPERIMENTAL PROCEDURES Startup. In total, the string of three cells were started three times with the CBS. The shared anolyte consisted of 1 g/L 5489

dx.doi.org/10.1021/es400239k | Environ. Sci. Technol. 2013, 47, 5488−5494

Environmental Science & Technology

Article

BES stack was stepped up through an identical series of galvanostatically applied currents.

■

RESULTS AND DISCUSSION Rapid Startup of Serially Stacked BESs. The CBS enable rapid startup of all BES cells in only 3 days following inoculation (results shown in Figure 2). With the CBS nodes

Figure 1. Block diagram of electrical connections in the experimental setup, including a CBS node simplified schematic. The arrows represent the direction of direct current in and out of the control system. The CBS nodes are connected to and controlled by PC-based software (not pictured). The power supply is controlled independent of the CBS.

sodium acetate in an autoclaved modified M9 medium as previously described.27 Each anode was filled with anolyte and inoculated with a microbial consortium from the effluent of an established, acetate-fed BES from an unrelated experiment.28 The effluent was circulated through the anode in open circuit for at least 3 days. After this open circuit period, the anode showed stable potentials below −0.2 V. At this point, a current of 50 mA was applied through the BES stack and a continuous feed to the anode was started. The anode chambers were individually fed at a flow rate of 0.7 L/day. The anode effluent from all cells was mixed and circulated at about 300 L/day using peristaltic pumps. All cathode chambers were fed from a shared 1 g/L sodium chloride solution at a flow rate of 1.4 L/ day. The cathode effluent was also mixed and circulated at 300 L/day. These flow rates and media remained consistent through the experiment. The anode chamber and current collector was thoroughly cleaned, and the graphite felt and cation-exchange membrane were replaced prior to each restart, because each control run (without CBS) was anode-destructive. Continuous Operation with Feed Stoppages. The BES stack was operated continuously with the CBS for 5 days after the productive current had stabilized. Over the following 7 days, each of the three cells was individually disconnected once from the feed supply for 24 and 48 h, then reconnected, and allowed a recovery period before the disconnection of the next cell. Following the recovery of the final BES cell, the stack was connected in series without the CBS nodes and a constant applied potential of 3.0 V, i.e., equal to the sum of the applied potential of the three BES cells when operated with the CBS nodes (set point of 1.0 V). After 4 days of continuous operation without the CBS nodes, the feed supply to a single BES cell was disconnected for 48 h and then reattached. Current Step-up Experiments. Current step-up experiments were performed on a stably operating BES stack that was started up with CBS nodes, as described above. The current was galvanostatically increased at values of 5, 10, 20, 40, and 80 mA for 1 h/step across the stack. The Tritium Pty Ltd. software tracked the current generated by individual BES cells according to the bypass percentage. The CBS was then removed, and the

Figure 2. (a) Current, (b) anode potential, and (c) applied cell voltage profile of three identical BES cells connected in series during startup, with and without CBS. With the CBS, the power supply galvanostatically supplies 50 mA to the BES stack and the control system sets the maximum individual BES cell voltage at 1.0 V (i.e., maximum BES stack voltage of 3.0 V). Without the CBS, the whole BES stack is controlled potentiostatically at 3.0 V and all individual currents are the same by the nature of a conventional serially connected BES stack. All data were logged once per minute as the average of 60 data points recorded once per second. Note that, in panel c for “CBS, n = 3,” σ = 0.

(cell voltage controlled at 1.0 V with 50 mA supplied current), the three BES cells in the stack started producing current with minimal intervention using a single power supply, which has not been reported previously in the literature. This is notable when one considers the current variation that can be seen in these identical reactors fed from a common source, as shown in Figure 2a. At t = 1.7 days, there was over 10 mA of variation in the system, generating an average of 10 mA. Stable performance was achieved at 15 mA after 2.5 days (average σ2 = 0.90 after t = 2.5 days). This stability was achieved by a consistent 5490

dx.doi.org/10.1021/es400239k | Environ. Sci. Technol. 2013, 47, 5488−5494

Environmental Science & Technology

Article

Figure 3. (a) Current and (b) anode potential profile of three identical serially connected cells controlled with (days 0−13, set cell voltage at 1.0 V, and 50 mA applied current) and without (days 13−20 and set stack voltage of 3.0 V) the CBS. The BES cells were maintained in continuous operation for approximately 4 days and then individually denied feed to simulate a performance trough. BES cell 1 had feed stopped for 24 h on day 4; BES cell 2 had feed stopped for 24 h on day 6; and BES cell 3 had feed stopped for 48 h on day 8. Without the CBS, BES cell 3 was denied feed for 48 h on day 16. In panel a for “No CBS, n = 3”, all individual currents are the same by the nature of a conventional serially connected BES stack. All data were logged once per minute as the average of 60 data points recorded once per second.

performance drop of an individual BES cell was not observed to affect the other two cells in the stack. In all cases, the disconnected BES cell fell to below 4 mA of productive current, with an associated rise in anode potential to around +0.15 V versus SHE. Simultaneously, the feed-connected BES cells continued generating around 25 mA of current and maintained anode potentials between 0.0 and −0.2 V versus SHE. Following a 24 h stoppage, BES cells 1 and 2 generated around 25 mA of current after their return to the feed source, while BES cell 3 only generated around 15 mA after a 48 h stoppage. After the CBS nodes were removed, the poorest performing BES cell was swiftly driven to a higher voltage but recovered and increased in current over the following 4 days. However, after the feed was stopped on day 16, the anode potential of this low-performing cell was driven to +1.30 V in 48 h, while the anode potentials of the other BES cells fell below −0.29 V. This resulted in an almost completely collapsed current across the stack ( 1.0 mA and (b) σ > 0.03 V versus SHE.

where the BES cells (n = 3) have consistent anode and stack potential with very low variation. From this mechanism, the CBS allows BES cell current generation independent of excess supplied current, demonstrated in Figure 4a, while panels b and c of Figure 4 highlight the protective nature of the CBS toward the anode biocatalysis. With the CBS, individual BES cells in a stack can draw a different amount of current relative to the activity of the bioanode and simultaneously bypass the surplus current back into the circuit. This process dynamically maintains the set voltage, thereby protecting the microorganisms at the bioanode from a destructive voltage, as seen in panels b and c of Figure 4. If the supplied current does not meet the productive capacity of a BES (e.g., less than 15 mA in panels a and b of Figure 4 at t < 2 h for all BES cells), the unit will show a cell voltage lower than the CBS set point of 1.0 V. Under these circumstances, the CBS will direct all of the applied current into a BES cell until it is running at maximum performance with respect to the applied current. As the supplied current increases, the BES cells reach a point where they cannot match the supplied current. When this occurs, the CBS control system maintains the set cell voltage by bypassing excess current back into the circuit, thereby allowing each cell to draw current while preventing its anode potential from increasing to meet the extra demand. Without the CBS, the anode potentials reached over +1.5 V versus SHE in all BES cells, as seen in Figure 4b. For applied currents of less than 20 mA, the anode potential profile without the CBS mirrors the profile with the CBS. This suggests that the three BES cells are indeed generating the applied currents 5492

dx.doi.org/10.1021/es400239k | Environ. Sci. Technol. 2013, 47, 5488−5494

Environmental Science & Technology

Article

Carrera et al.9 A related challenge in applied BESs is the fact that resistance across the membrane will change over time because of membrane degradation, precipitation, and fouling, requiring a higher applied potential to achieve the same current.30 This applied potential also needs to be managed relative to the potential and conditions of both the bioanode and the cathode, such as pH, residence time, etc. Such problems are compounded by the difficulty of using sensitive reference electrodes to monitor electrolyte potentials in industrial applications, particularly in streams with suspended solids. Feedback from the CBS on the cell voltage and bypass percentage (i.e., individual BES cell current) can provide information for optimization of inputs relative to individual BES cell performance, thereby improving power and feed utility and overall efficiency. Finally, an important consideration in applied biotechnology is unit remediation. In the event of a BES cell failure, the CBS control system diverts all current to the stack circuit for the continued operation of the remainder of the stack. Therefore, any individual cell can be removed from the stack and replaced without significantly affecting the performance of other cells and the performance of the stack overall. To step from large-scale systems back to the laboratory, it is useful to finally note that the CBS may be of high value for single BESs in terms of startup in general practice. Indeed, as shown here, the CBS-controlled BESs started up rapidly likely because of the fact that current was not limited by the system (a higher set point than strictly required was always applied) and potential could simultaneously evolve to a level where the cell is most functional. This fact makes the CBS attractive also for laboratory reactors, where rapid startup of a novel process is desired without predefined knowledge in the process rate or optimal potential.

■

(2) Logan, B. E. Simultaneous wastewater treatment and biological electricity generation. Water Sci. Technol. 2005, 52 (1−2), 31−37. (3) Pant, D.; Singh, A.; Van Bogaert, G.; Irving Olsen, S.; Singh Nigam, P.; Diels, L.; Vanbroekhoven, K. Bioelectrochemical systems (BES) for sustainable energy production and product recovery from organic wastes and industrial wastewaters. RSC Adv. 2012, 2 (4), 1248−1263. (4) Harnisch, F.; Aulenta, F.; Schröder, U. 6.49Microbial fuel cells and bioelectrochemical systems: Industrial and environmental biotechnologies based on extracellular electron transfer. In Comprehensive Biotechnology, 2nd ed.; Moo-Young, M., Ed.; Academic Press: Waltham, MA, 2011; Vol. 6, pp 643−659. (5) Pant, D.; Singh, A.; Van Bogaert, G.; Gallego, Y. A.; Diels, L.; Vanbroekhoven, K. An introduction to the life cycle assessment (LCA) of bioelectrochemical systems (BES) for sustainable energy and product generation: Relevance and key aspects. Renewable Sustainable Energy Rev. 2011, 15 (2), 1305−1313. (6) Rabaey, K.; Bützer, S.; Brown, S.; Keller, J.; Rozendal, R. A. High current generation coupled to caustic production using a lamellar bioelectrochemical system. Environ. Sci. Technol. 2010, 44 (11), 4315− 4321. (7) Wang, X.; Cai, Z.; Zhou, Q.; Zhang, Z.; Chen, C. Bioelectrochemical stimulation of petroleum hydrocarbon degradation in saline soil using U-tube microbial fuel cells. Biotechnol. Bioeng. 2012, 109 (2), 426−433. (8) Kim, Y.; Logan, B. E. Microbial desalination cells for energy production and desalination. Desalination 2013, 308, 122−130. (9) Gil-Carrera, L.; Escapa, A.; Mehta, P.; Santoyo, G.; Guiot, S. R.; Morán, A.; Tartakovsky, B. Microbial electrolysis cell scale-up for combined wastewater treatment and hydrogen production. Bioresour. Technol. 2013, 130, 584−591. (10) Wang, L.; Chen, Y.; Huang, Q.; Feng, Y.; Zhu, S.; Shen, S. Hydrogen production with carbon nanotubes based cathode catalysts in microbial electrolysis cells. J. Chem. Technol. Biotechnol. 2012, 87 (8), 1150−1156. (11) Kim, Y.; Hatzell, M. C.; Hutchinson, A. J.; Logan, B. E. Capturing power at higher voltages from arrays of microbial fuel cells without voltage reversal. Energy Environ. Sci. 2011, 4 (11), 4662−4667. (12) Meehan, A.; Hongwei, G.; Lewandowski, Z. Energy harvesting with microbial fuel cell and power management system. IEEE Trans. Power Electron. 2011, 26 (1), 176−181. (13) Donovan, C.; Dewan, A.; Peng, H.; Heo, D.; Beyenal, H. Power management system for a 2.5 W remote sensor powered by a sediment microbial fuel cell. J. Power Sources 2011, 196 (3), 1171−1177. (14) Grondin, F.; Perrier, M.; Tartakovsky, B. Microbial fuel cell operation with intermittent connection of the electrical load. J. Power Sources 2012, 208, 18−23. (15) Wang, H.; Park, J.-D.; Ren, Z. Active energy harvesting from microbial fuel cells at the maximum power point without using resistors. Environ. Sci. Technol. 2012, 46 (9), 5247−5252. (16) Dewan, A.; Beyenal, H.; Lewandowski, Z. Intermittent energy harvesting improves the performance of microbial fuel cells. Environ. Sci. Technol. 2009, 43 (12), 4600−4605. (17) Wu, P. K.; Biffinger, J. C.; Fitzgerald, L. A.; Ringeisen, B. R. A low power DC/DC booster circuit designed for microbial fuel cells. Process Biochem. 2012, 47 (11), 1620−1626. (18) Kutkut, N. H.; Divan, D. M. Dynamic equalization techniques for series battery stacks. Proceedings of the 18th International Telecommunications Energy Conference, INTELEC ’96; Boston, MA, Oct 6−10, 1996; pp 514−521. (19) Aelterman, P.; Rabaey, K.; Pham, H. T.; Boon, N.; Verstraete, W. Continuous electricity generation at high voltages and currents using stacked microbial fuel cells. Environ. Sci. Technol. 2006, 40 (10), 3388−3394. (20) Kim, D.; An, J.; Kim, B.; Jang, J. K.; Kim, B. H.; Chang, I. S. Scaling-up microbial fuel cells: Configuration and potential drop phenomenon at series connection of unit cells in shared anolyte. ChemSusChem 2012, 5 (6), 1086−1091.

AUTHOR INFORMATION

Corresponding Author

*Telephone: +32-9-264-59-76. Fax: +32-9-264-62-48. E-mail: [email protected]. Notes

The authors declare no competing financial interest. # Korneel Rabaey and René A. Rozendal assume senior leadership.

■

ACKNOWLEDGMENTS This research was supported by Bilexys Pty Ltd. and the Queensland Sustainable Energy and Innovation Fund (BNE42153). At Ghent University, Stephen J. Andersen is funded by the European Community’s Seventh Framework Programme FP7/2007-2013 under Grant Agreement 226532. Korneel Rabaey is supported by the European Research Council (ERC Starter Grant ELECTROTALK). The authors thank Balavelan Thanigaivelan for the preliminary work performed during his industry internship funded by the Australian Mathematical Sciences Institute (AMSI). The authors acknowledge and appreciate the ongoing support of James Kennedy and Tritium Pty Ltd. The authors thank Tim Lacoere for the graphical art.

■

REFERENCES

(1) Rabaey, K.; Rozendal, R. A. Microbial electrosynthesis Revisiting the electrical route for microbial production. Nat. Rev. Microbiol. 2010, 8 (10), 706−716. 5493

dx.doi.org/10.1021/es400239k | Environ. Sci. Technol. 2013, 47, 5488−5494

Environmental Science & Technology

Article

(21) Oh, S. E.; Logan, B. E. Voltage reversal during microbial fuel cell stack operation. J. Power Sources 2007, 167 (1), 11−17. (22) Dekker, A.; Ter Heijne, A.; Saakes, M.; Hamelers, H. V. M; Buisman, C. J. N. Analysis and improvement of a scaled-up and stacked microbial fuel cell. Environ. Sci. Technol. 2009, 43 (23), 9038− 9042. (23) Shin, S. H.; Choi, Y.; Na, S. H.; Jung, S.; Kim, S. Development of bipolar plate stack type microbial fuel cells. Bull. Korean Chem. Soc. 2006, 27 (2), 281−285. (24) Aelterman, P.; Rabaey, K.; Pham, H. T.; Boon, N.; Verstraete, W. Continuous electricity generation at high voltages and currents using stacked microbial fuel cells. Environ. Sci. Technol. 2006, 40 (10), 3388−3394. (25) Baliga, B. J. Fundamentals of Power Semiconductor Devices; Springer: New York: 2008; pp 276−503. (26) Tritium Pty Ltd. IQ Battery Management System; Tritium Pty Ltd.: Tennyson, Queensland, Australia; http://tritium.com.au/ products/iq-battery-management-system/. (27) Rabaey, K.; Ossieur, W.; Verhaege, M.; Verstraete, W. Continuous microbial fuel cells convert carbohydrates to electricity. Water Sci. Technol. 2005, 52 (1−2), 515−523. (28) Guo, K.; Chen, X.; Freguia, S.; Donose, B. C. Spontaneous modification of carbon surface with neutral red from its diazonium salts for bioelectrochemical systems. Biosens. Bioelectron. 2013, 47, 184−189. (29) Tartakovsky, B.; Mehta, P.; Santoyo, G.; Guiot, S. R. Maximizing hydrogen production in a microbial electrolysis cell by real-time optimization of applied voltage. Int. J. Hydrogen Energy 2011, 36 (17), 10557−10564. (30) Choi, M. J.; Chae, K. J.; Ajayi, F. F.; Kim, K. Y.; Yu, H. W.; Kim, C. W.; Kim, I. S. Effects of biofouling on ion transport through cation exchange membranes and microbial fuel cell performance. Bioresour. Technol. 2011, 102 (1), 298−303.

5494

dx.doi.org/10.1021/es400239k | Environ. Sci. Technol. 2013, 47, 5488−5494