Enhancement of Electricity Generation by a Microbial Fuel Cell Using

Subscriber access provided by University of Newcastle, Australia

Article

Enhancement of Electricity Generation by a Microbial Fuel Cell Using a Highly Active Non-Precious Metal Nitrogen-Doped Carbon Composite Catalyst Cathode Gregory Ryan Dong, Hamid-Reza Kariminia, Zhongwei W. Chen, Wayne J. Parker, Mark D. Pritzker, and Raymond L. Legge Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.6b02206 • Publication Date (Web): 12 Dec 2016 Downloaded from http://pubs.acs.org on December 15, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Energy & Fuels is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Enhancement of Electricity Generation by a Microbial Fuel Cell Using a Highly Active NonPrecious Metal Nitrogen-Doped Carbon Composite Catalyst Cathode Gregory Ryan Dong, Hamid-Reza Kariminia§, Zhongwei W. Chen, Wayne Parker, Mark D. Pritzker and Raymond L. Legge*

Department of Chemical Engineering, University of Waterloo, 200 University Avenue West, Waterloo, ON, N2L 3G1, Canada § Department of Chemical and Petroleum Engineering, Sharif University of Technology, Tehran, P.O. Box 11155-9465, Iran

KEYWORDS: oxygen reduction; non-precious metal catalyst; microbial fuel cell; bioelectricity generation

1 Environment ACS Paragon Plus

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT As microbial fuel cell (MFC) technology continues to gain momentum towards commercialization, the replacement of traditionally used platinum for oxygen reduction with an inexpensive catalyst becomes more important. A non-precious nitrogen-doped carbon composite catalyst with previous applications in PEM fuel cells is demonstrated for the first time in a single-chamber air cathode MFC with comparisons to a similar platinum-based MFC. The performance of the MFC is compared with a similar MFC using a platinum catalyst and acetate feed. When the platinum is replaced with the catalyst loaded at the surface of the proton exchange membrane (loading density of 1 mg/cm2), MFC operation outperforms a similar platinum-based cell. The synthesized catalyst produced 213.2 ± 13.9 mW/m2 power density that on average is 151% higher than that for the platinum catalyst. Columbic efficiency was also higher as a result at 6.71 ± 0.88 %. A two-fold increase in the loading density (2 mg/cm2) for the synthesized catalyst resulted in a 305% increase in the generated power compared to the platinum catalyst. This suggests that the non-precious nitrogen-doped carbon composite is a potentially attractive replacement for conventional platinum as a catalyst for energy production by MFCs.

1. INTRODUCTION Microbial fuel cells (MFCs) are a promising technology that combines the generation of electricity with the treatment of wastewater through the metabolic oxidation of organic and inorganic substrates by bacterial species living within a biofilm. The origin of using microorganisms to generate electricity is not a new concept and dates back to 1910 when Potter


Page 2 of 35

Page 3 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

first discovered electricity production by E.coli.1 These results were later substantiated in 1931 when Cohen demonstrated that a voltage of 35 V could be achieved from MFCs connected in series.1 Although some studies in the 1950s and 1960s were conducted, the past 20 years have seen a significant push in the area of MFC research. A key aspect for the successful commercialization of MFCs is not only their power production, but also their material costs. Perhaps the most significant capital cost of an air-cathode MFC is that associated with the catalyst. Platinum has traditionally been used as the catalyst in dual chamber proton exchange and earlier types of MFCs2,3 and its use has crossed over to different types of MFCs as well. Platinum is typically used as the cathode catalyst in air-cathode MFCs as it is considered the best known catalyst for oxygen reduction.6 One of the reasons for this preference is due to its high oxygen reduction reaction exchange current density (2.8 × 10-3 mA/cm2). Platinum in fact has the highest exchange current densities of any metal, making it a suitable choice.4 Another advantage of platinum is that it has a high selectivity for the 4-electron reduction pathway to water (99.5%), as opposed to the 2-electron reduction pathway to hydrogen peroxide (0.5%).5 The primary disadvantage of using a platinum catalyst in any application is cost. Since MFCs to date generate lower power densities than proton exchange MFCs, the cost of platinum (per unit power) is even more significant. Therefore, new non-precious catalysts which can produce similar power densities at a fraction of the cost of platinum are required to improve the attractiveness of MFCs as a practical method of generating power on a commercial scale. The development of such a catalyst may ultimately determine the financial viability of MFCs in the future. Some studies have found that non-precious cathode catalysts such as cobalt tetramethoxyphenylporphyrin (CoTMPP) and iron phthalocyanine can achieve similar or even improved fuel cell performance over traditional platinum catalysts.7,8 HaoYu et al. compared several different non-precious


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

catalysts such as pyrolyzed iron phthalocyanine (FePc), CoTMPP, and manganese phthalocyanine (MnPc). They found that FePc (634 mW/m) outperformed the other catalysts, including platinum (593 mW/m2), with comparable coulombic efficiencies of ~20%. They also compared the electrochemical performance of FePc using two different carbon black substrates, Vulcan XC-72 and KetjenBlack EC 300J. KetjenBlack was found to have a higher OCP (0.319 V vs. 0.289 V (vs. Ag/AgCl electrode)). Although no further testing was done to determine the exact reason behind this increase, it was suspected that the increased surface area of the KetjenBlack allowed for more adsorption of oxygen to the catalyst/carbon surface.7 Activated carbon fiber felt without a metal catalyst has been successfully used as a cathode in MFCs by Cheng et al. who utilized ~0.6 mg/cm2 cobalt tetramethoxyphenylporyphyrin as a catalyst that produced only 12% less power than a 0.5 mg/cm2 Pt catalyst (369 vs. 414 mW/m2) at similar coulombic efficiencies of 8-18%.9 These findings clearly give promise that inexpensive nonprecious catalysts can be feasible alternatives to platinum. Some new approaches to catalyst development for fuel cell applications have focused not only on shifting from precious metal catalysts, but also on incorporating nitrogen groups into the carbon support to make the support itself catalytic for oxygen reduction. By so doing, significantly more surface area is made available for oxygen reduction than the traditional incorporation of metal catalyst onto the surface of a carbon support, which does not participate in oxygen reduction. Different types of nitrogen-doped carbon-based electrodes have been synthesized and utilized in MFCs.10-15 Ethylenediamine has been used in the anode electrode of MFCs for modification of graphite felt16 or polymerized polyaniline.17 There are only a few studies in the literature on the performance of MFC using a cathode catalyst fabricated using ethylenediamine or ethylenediaminetetraacetic as a nitrogen precursor have been reported.18-22 Nguyen fabricated a


Page 4 of 35

Page 5 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

cathode catalyst containing iron chelates of ethylenediamine group that was tested in a MFC.23 These catalysts have been widely used in proton exchange membrane (PEM) fuel cells to enhance the oxygen reduction reaction.24 Choi et al. reported on the fabrication and characterization of a novel ethylenediamine-based non-precious metal-nitrogen catalyst for oxygen reduction and evaluation of its application in PEM chemical fuel cells.25 In this work, we examine the application and performance of this catalyst for the first time in a single chamber air cathode MFC. The focus of this study is the use and evaluation of an iron/cobalt ethylenediamine-based nitrogen-doped carbon composite catalyst for cathodic oxygen reduction in a single chamber MFC. Conclusions based on comparisons of the overall current, power density and chemical oxygen demand (COD) removal to that obtained using a similar platinum-based cell are presented.

2. MATERIALS AND METHODS 2.1. Catalysts used. The non-precious metal catalyst was fabricated according to a procedure described previously25 except that the support was not pre-treated in this work. A non-precious catalyst was made using ethylenediamine as a nitrogen-containing precursor and the support was a highly porous carbon (Ketjen Black EC-600JD). After sonication and mixing with ethanol (125 mL), the support was added to a solution containing equal masses of cobalt nitrate, iron sulfate and ethylenediamine (2mL) dissolved in 125 mL ethanol. The resulting mixture was boiled to dryness and then the residue was ground and pyrolyzed in argon gas at 900 ºC for 1 hour before being acid-leached in 0.5 M H2SO4 (for 8 h) to remove the transition metals. Finally, the


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

remaining solid catalyst was washed, filtered and dried before storage. For the experiments with platinum as the catalyst, 10% Pt on the activated charcoal support (VULCAN-XC 72) from Sigma-Aldrich was used. A flowchart showing different stages for fabrication of the catalyst is given in Supporting Information Figure S1.

2.2. Measurements and Analyses. SEM images were obtained using a LEO FESEM 1530 electron microscope. The sample was prepared by airbrushing a pre-made catalyst ink onto a piece of carbon tape adhering to a metallic holder. Powder X-ray diffraction (XRD) analysis was used to identify the crystalline compounds present in the catalyst (PANalytical Xpert Pro Materials Research Diffractometer), while energy-dispersive X-ray spectroscopy (EDAX) was used to determine its elemental composition (LEO1530 FE-SEM).

During MFC operation, samples were taken periodically from the anolyte and their pH, oxidation-reduction potential (ORP), dissolved oxygen (DO), total suspended solid (TSS)/volatile suspended solid (VSS) and total and soluble COD content were measured according to standard methods.26 The MFCs were fed with a known COD concentration (6.44 g/L) in the form of sodium acetate. The electrical charge generated is proportional to a COD equivalence based on the oxygen reduction reaction (see eq 3). Total COD (tCOD) and soluble COD (sCOD) of both the feed and discharge were measured every 2 days to determine the amount of suspended biomass produced, while head gas samples were collected in Tedlar® bags to measure the gas volume using the water displacement method and analyzing its composition every 4-6 days by gas chromatography (GC) for methane composition. Before being attached to


Page 6 of 35

Page 7 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

the head gas port of the MFC, the bag was emptied of any residual gas using a vacuum pump and locked using the adjustable nipple. The amount of methane produced was then converted into a COD-equivalent. Based on these COD-equivalents, the coulombic efficiency could be determined and an account of the COD balance for the entire process made. As a check of the soluble COD measurements, the acetate concentration in the feed and discharge was also directly obtained by IC analysis.

Acetate concentration was analyzed by ion chromatography using a Dionex DX-300 chromatography system (Sunnyvale, CA, USA) equipped with a UV/Vis detector using a 2-20% acetonitrile and 0.005 M 1-methanesulfonic acid eluent. Gas samples from the Tedlar® bags (Chromatographic Specialties, Brockville, ON, Canada) were collected once every 4 days and analyzed by GC (SRI 310C Gas Chromatograph, TCD detector, helium gas carrier) to measure methane, nitrogen/oxygen (the two are indistinguishable from each other for this GC setup) and carbon dioxide fractions of the sample.

All electrochemical parameters were recorded using the eDAQ eCorder Model 401 data logger. The raw data collected included the current and cell voltage. Using the average current (I) measured over time, the coulombic charge Q generated over the time interval from

to

was

determined from eq 1:

(1)

The power P generated by the cell was calculated from the cell voltage and current as follows: 7 Environment ACS Paragon Plus

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 35

(2) Since the cell voltage was held constant at 0.3 V throughout this study, the power was directly proportional to the current. The COD equivalence of the generated current was calculated as shown in eq 3 for a fed-batch reactor.

(3)

where CODElec is the COD equivalence of electricity (g/L) generated in the MFC, F is the Faraday constant (96485 C/mol),

denotes the liquid volume (L) in the anolyte chamber and

the factor of 8 arises from the ratio of the molecular mass of oxygen (32 g/mol O2) to the number of moles of electrons generated per mole of oxygen (4 mol e-/mol O2). Once the CODElec was found, the coulombic efficiency (CE), which is an indication of MFC ability to convert an available substrate to current, was then determined as follows:

(4) where ∆COD (g/L) = sCODFeed – sCODDischarge ). The soluble COD was used to determine

∆COD since only the conversion of the main substrate acetate was of interest in this study.

2.3. Microbial Fuel Cell Setup. The microbial fuel cell was constructed as a two-cell stack with a flat-plate design similar to the dual-chamber flat-plate cell. In this design, no catholyte compartment was used and the cathode was exposed directly to ambient air. Each anolyte


Page 9 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

chamber, which held the wastewater feed stream, contained an inlet line, outlet line and baffles designed to help prevent short-circuiting of the flow and to increase the tortuosity of the flow path through the reactor. The design consisted of two plain carbon paper anodes and a nonconductive plate placed between them to electrically and physically separate the two cells. On the other side of the two cells were the cathodes. The cathodes consisted of single-sided membrane-electrode assemblies (MEA). Each MEA was comprised of two layers: i) carbon paper containing a hydrophobic PTFE layer (5% weight) to reduce water loss, microporous gas diffusion layer (GDL) to improve oxygen diffusion and a home-made Pt-coated or non-precious ethylenediamine-based carbon composite catalyst airbrushed onto the GDL, and ii) a pre-treated sheet of the proton-exchange membrane (Nafion). The Nafion sheet was pre-treated by boiling in 3% hydrogen peroxide for 1 h to remove organic contaminants, rinsing with DI water for 1 h, boiling in 0.5 M H2SO4 acid for 1 h to convert from the Na-form to H-form of the ionomer and finally rinsing for another 1 h in DI water to remove any free acid remaining in the membrane. The two layers of the MEA were annealed by hot-pressing at 1780 kPa and 140ºC for 3 min.27

The connections of the cell to the external circuit were made directly at the carbon paper anode and at a stainless steel mesh pressed to the cathode using a perforated stainless steel plate. All liquid-containing parts of the cell were sealed using rubber gaskets. The MFC stack as a whole was constructed as a plate-and-frame design fastened together by threaded Nylon screws. The overall nominal flat-plane area of anode and cathode was both 80 cm2 and the overall anolyte chamber volume was 120 mL. In a series of experiments, MFC-EDA-1 and MFC-EDA-2 operated with a non-precious metal catalyst at the loading densities of 1 and 2 mg/cm2,


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 35

respectively. Photographs of the baffled anolyte chamber along with the fuel cell (shown as the two-cell stack) are presented in Figure 1. Figure 1

The MFC was operated in an incubator at 30 ºC. The inlet stream was fed from an external chiller controlled at 4 ºC to reduce microbial degradation of the substrate. The feed system consisted of a peristaltic pump controlled by two remote timers for switching the pump on or off. Tedlar bags connected downstream from the MFC enabled any gas generated by the process to be collected and analyzed. A potentiostat was used to control the cell potential and measure the current generated by each cell. The charge was measured by a digital coulometer connected to the potentiostat, converted into a current signal and sent to the data logger. The analog output measurements from the potentiostat were then converted by an eDAQ E-corder Model 401 data logger into a digital signal that was relayed to a PC computer for real-time data monitoring and collection.

2.4. Feed Solution and System Operation. A mixture of the digester overflow and aeration basin collected from the Waterloo domestic wastewater treatment plant was used as an inoculum. Feed conditions were chosen based on the conditions that yielded the highest power production during the preliminary experiments with the platinum catalyst (data not shown). The synthetic feed solution consisted of acetate (nutrient source), 5 mM phosphate buffer and the following minerals (in g/L as NH4Cl, 1.073; FeCl3·6H2O, 0.030; MgCl2·6H2O, 0.644; CoCl2, 0.054; ZnSO4, 0.029; CuSO4, 0.027; CaCl2, 0.011; MnCl2·4H2O, 0.051). The final feed solution 10 Environment ACS Paragon Plus

Page 11 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

was titrated to pH 7 using sodium hydroxide. All experiments were conducted in an incubator set at 30ºC at a controlled potential of 0.3 V between the anode and cathode. This potential was chosen based on previous MFC experiments done in our group.28 This controlled voltage was chosen as an operational midpoint between MFC short circuit (V=0) and an average of the open circuit potential of the two MFCs tested before (average V≈0.6). Operation of the fuel cell was carried out at a COD concentration of 6.44 g/L as acetate, HRT of 24 h, feed frequency of 6 cycles/day (each cycle contained 20 mL feed solution). The effective volume of the anolyte contained within the cell was 120 mL. The feeding rate was set to 20 mL/cycle. After inoculation for 5 days, the feed solution was adjusted to the acetate substrate. Cells were operated until the average daily current density varied by less than 5%. For MFC-EDA-1, stabilization occurred after 19 days while MFC-EDA-2 required 29 days.

3. RESULTS AND DISCUSSION

3.1. Characterization of the nitrogen-doped catalyst. The nitrogen-doped ethylenediamine-based catalyst was synthesized and its characteristics determined according to the materials and methods.

3.1.1. SEM Analysis. A sample of the EDA-based catalyst was examined by SEM. An image obtained at 10,000x magnification is presented in Figure 2a. The surface of the catalyst appears to be fluffy, indicative of a highly porous structure. This is expected as the catalyst is derived from a carbon black powder. A previous study found that the structure of a similar


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 35

catalyst synthesized using polyaniline was fibrous prior to pyrolysis, but became more bulbous after pyrolysis.29. As seen in the 100,000x inset, the structure of the catalyst appears bulbous.

3.1.2. EDAX/XRD Analysis. From EDAX, the elemental composition of the catalyst sample consists mostly of carbon, as expected. The other elements detected were oxygen (4.9 wt %), nitrogen (3.6 wt %), sulfur (1.7 wt %), iron (1.3 wt %) and cobalt (1.3 wt %). Nitrogen is incorporated into the carbon structure through the EDA precursor, while the transition metals (Fe and Co) are likely bound to the nitrogen groups, which in turn are bound to carbon. The metalnitrogen-carbon catalysts can be divided into two structure types: pyrrole-like (MeN4/C, where Me is the transition metal) and pyridinic-like (MeN2/C) structures. The iron content in our catalyst was similar to that reported by Nallathambi et al.30 by ICP-mass spectrometry measurements (i.e., 1.4 wt %), although the cobalt content was found to be significantly higher (4.6 wt %). The oxygen content on an atomic basis was slightly higher than that of nitrogen. The presence of sulfur in the catalyst can likely be attributed to the use of sulfuric acid during the leaching process.

Determination of the crystalline structure of the catalyst was carried out by XRD. The diffractogram obtained for the EDA catalyst sample is shown in Figure 2b. All peaks were identified based on the Powder Diffraction File 2.0 provided by the International Centre for Diffraction Data. From Figure 2b, the most intense peak at 46.79º (denoted by +) is attributed to either Fe, Co or FexCoy, while the less intense peaks are assigned to Co at 51.45º and Fe at 64.38º and 81.58º. The broad peak at about 25º is indicative of the carbon composite catalyst,


Page 13 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

while the sharp peak at 40.11º can be attributed to aluminum in the sample dish. A small unknown peak was detected at approximately 31º, possibly due to an organic or organometallic compound that may have formed during synthesis. A peak at 45.6º that is partially confounded by the more intense peak at 46.79º also appears. In their earlier study, Nallathambi et al. also observed this peak and assigned it to Fe3C (cementite). They also found peaks similar in intensity to the “+” peak (for Fe, Co or FexCoy) and the “o” peak (carbon composite) appearing in Figure 2b.30 The detection of Fe and Co by both EDAX and XRD is not surprising given the synthesis procedure. The detection of these metal compounds by XRD provides more evidence that these compounds were present in the catalyst. XRD of the carbon support (Ketjen Black EC-600JD) has been compared with the catalyst where elsewhere25. The carbon support shows a distinct peak at 2θ=24.3° only.

Figure 2

3.1.3. Other Specifications of the catalyst. Other specifications of the catalyst have been studied previously. The carbon support characteristics was previously determined against the synthesized catalyst. The BET surface area of the carbon support and catalyst was 483.7 and 1416.2 m2/g, respectively. This was higher than the BET surface area of the commercially available support for the platinum catalyst used in this work (∼250 m2/g).31 The pore volume of the carbon support and catalyst was 2.279 and 0.541 cm3/g, respectively. According to former work, TEM images showed that the average size of the platinum nanoparticles were 3 to 4 nm and the synthesized catalyst had a stable catalyst structure32. Linear sweep voltammetry of the


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 35

catalyst and platinum had been previously conducted. A high degradation of the Pt/C was observed in comparison with the synthesized catalyst that was proved by a decrease in the oxygen reduction reaction performance. Pt/C catalyst could only retain 32% of its initial mass activity in accelerated degradation tests; however, the synthesized catalyst retained as much as 80% of its initial activity.32 It appears that the 4-electron transfer oxygen reduction reaction (ORR) is the dominant mechanism in the synthesized catalyst. After adsorption of oxygen to the catalyst surface, it can either go through 2-electron transfer ORR that generates hydrogen peroxide, or 4-electron transfer ORR that produces water. A mixture of these two pathways occurs in practice.33 It has been shown that the fractional yield of hydrogen peroxide in the synthesized catalyst is 5-10%25 which corroborates the domination of 4-electron transfer pathway.

3.2 Performance of EDA-based catalyst in microbial fuel cell. The synthesized nitrogen-doped catalyst was utilized as the cathode for oxygen reduction in a single chamber-air cathode -MFC. The performance of the MFC using this catalyst was evaluated at loading densities of 1 and 2 mg/cm2 (denoted MFC-EDA-1 and MFC-EDA-2, respectively) and compared to that obtained in the same MFC using platinum as a catalyst (denoted MFC-Pt). The properties measured included the current production, pH, ORP, DO, solids analysis (TSS/VSS), methane production, COD measurement balance and acetate concentration for operation of the MFC at 0.3 V. For purposes here, MFC performance was considered to have reached a stable state when the variance of the current production over a period of 6 HRTs (6 days) was less than ~ 5%. A variation of the current densities from both MFC-EDA-1 and MFC-EDA-2 as well as


Page 15 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

MFC-Pt is given in Figure 3. MFC-EDA-2 required a longer time to reach a stable condition with a higher current density. Maximum current density for MFC-Pt under stable conditions was less than one third of that for MFC-EDA-2.

Figure 3

3.2.1. Power generation in MFC-EDA. As seen from Table 1, the current density obtained from both cells at 0.3 V using the non-precious metal catalyst exceeds the platinum cell by significant margins, an increase of 50.5% in MFC-EDA-1 and 205% in MFC-EDA-2. Another observation is that the current density generated by MFC-EDA-2 was almost exactly double that by MFC-EDA-1, which corresponds to the ratio of their catalyst loadings. This is an indicator that the rate of the cathodic reaction was in fact the limiting factor for current generation. Since the system appears to be operating under conditions where current generation is linearly proportional to the catalyst loading, even higher current and power could presumably be generated by the MFC if the catalyst loading were increased further. From the point of view of catalyst morphology, a large portion of the surface area of the N-doped carbon catalyst nanoparticles could participate in oxygen reduction since several different types of sites may be active in the surface structure such as the edge planes and defects as well as transition metal species.34-36 In the platinum catalyst, only the exposed surface area of the platinum particles is available for oxygen reduction, while the carbon support has negligible oxygen reduction capability. Although platinum can likely catalyze oxygen reduction to a greater extent than N-


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 35

doped carbon when normalized with respect to the catalyst area, its overall active surface area is likely much lower than that of the N-doped carbon catalyst used here.

Table 1

The polarization and power curves obtained using MFC-EDA-2 are compared to those obtained by MFC-Pt in Figure 4. The internal resistances in MFC-Pt and MFC-EDA-2 were 37.4 and 13.7 Ω, respectively. Experimental problems at the end of the cycle for MFC-EDA-1 led to inactivation of the biofilm and a polarization curve was not obtained for this cell. However, by considering the measured current of this cell, the internal resistance can be estimated to be at a value between of that for MFC-Pt and MFC-EDA-2. Maximum power density for MFC-EDA-2 and MFC-Pt was, 568.7 and 213.8 mW/cm2, respectively. Open circuit potential (OCP) of both cells were near 0.6 V. In the application of the cathode catalyst fabricated using ethylenediamine as a nitrogen precursor utilized in MFC, different OCPs has been reported (over 600 mV18, 780 mV19, and near 400 mV20). A higher OCP is an indication of higher reaction rate. In a MFC, the electromotive force of 1.1 V is the maximum potential that can be achieved, theoretically. Losses associated with MFC characteristics including overpotentials of the anode and the cathode as well ohmic losses are the cause of lower voltages in practice.37 The superiority of a similar nonprecious metal catalyst overplatinum has been observed in PEM fuel cells. The EDA-based catalyst showed a peak power density which was 67% higher than that for the Pt catalyst.32


Page 17 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 4

3.2.2. Methane Production. The head gas generated was collected in Tedlar® bags and its volume measured every 4-6 days using liquid displacement and then analyzed for methane, carbon dioxide and nitrogen/oxygen by GC. The variation of the production rate of each gas and the total amount is shown in Figure 5 while a summary of the gas production rate during the stable periods is given in Table 1. In anaerobic digesters, the doubling time of methanogens can be on the order of 1-9 days due to their slow growth.38 Even in an unmixed system such as the MFC where biomass can accumulate, the HRT of 24 h may be long enough for the population of methanogens in the MFC to grow significantly. The average gas production rate of 176.8 ± 37.6 mL/day from MFC-EDA2 during the stable period was higher than from both MFC-EDA-1 and MFC-Pt during a comparable period. Although the current generated by MFC-EDA-2 stabilized 8 days later than it did in MFC-EDA-1, MFC-EDA-2 was already producing nearly 200 mL/day of gas at day 15, the beginning of the stable region for MFC-EDA-1, whereas MFC-EDA-1 was producing ~134 ml/day. A similar difference in acetate COD removal by MFC-EDA-1 and MFC-EDA-2 was found. A study by Virdis et al.39 found that a higher proportion of available COD was converted into methane at a controlled anodic potential of −200 mV than at an anodic potential of −100 mV (40.1% vs. 28.8%), while the current produced was similar. The amount of biomass produced also declined as the anodic potential became more negative. However, Virdis et al. did not report the cathode or cell potential and power produced for their study. In the current study the cell voltage and not the anode potential was controlled; it is possible that the anodic potential of


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 35

MFC-EDA-2 may have been lower than that of MFC-EDA-1 so that more energy was available for growth and methane production. It was also found that the CO2/CH4 ratio in the gas stream discharged from both MFC-EDA-1 and MFC-EDA-2 was much higher than that from MFC-Pt. It is known that methanogens can also metabolize carbon dioxide to methane. Thus, the difference in the CO2/CH4 ratios suggests that the population of methanogens in MFC-EDA-1 and MFC-EDA-2 may have been lower than in MFC-Pt, decreasing the conversion of carbon dioxide to methane.

Figure 5

3.2.3. Variation of COD. Samples of the feed stream and the discharge from MFC-EDA-1 and MFC-EDA-2 were analyzed for total, soluble COD and acetate and are presented in Figure 6. A summary of the average tCOD, sCOD and acetate concentrations over the period of stable operation are presented in Table 1. Both cells were effective at degrading acetate from the feed, with MFC-EDA-1 removing 79.6% and MFC-EDA-2 removing 92.2% of soluble COD. Unlike the situation with MFC-Pt, the total COD concentrations in the discharge from each cell differed significantly from the soluble COD level (Table 1). Visual inspection of the discharge streams from both cells showed that they were generally darker than those produced in MFC-Pt, indicating the presence of biomass. This finding was confirmed through the TSS and VSS analyses that showed higher levels than for the MFC-Pt. At times, large amounts of biomass were found in samples collected from MFC-EDA2, as well as smaller amounts in several samples from MFC-EDA-1. The discharge from MFC18 Environment ACS Paragon Plus

Page 19 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

EDA-1 contained significantly more total and soluble COD than did the discharge from MFCEDA-2, amounting to differences of 1.06 g tCOD/L and 0.82 g sCOD/L, respectively. Although MFC-EDA-1 had been operating for less time than MFC-EDA-2 when its current production had stabilized (8 days less), a clear difference in the COD removal rate was evident by day 13. It is possible that the faster COD removal in MFC-EDA-2 may have been due to the anodic potential being lower in MFC-EDA-2 than MFC-EDA-1, permitting increased methane generation and biomass growth. It should be noted though that the discharge tCOD and sCOD levels in the discharge from MFC-EDA-1 were still decreasing at the end of the run. Given more time, COD removal from MFC-EDA-1 may have increased further.

Figure 6

Based on the soluble COD for the feed and discharge, the average coulombic efficiencies were calculated and compared (Table 1). Both MFC-EDAs exhibited improved coulombic efficiencies over those obtained for MFC-Pt, with MFC-EDA-1 and MFC-EDA-2 attaining levels of 6.71% and 12.18%, respectively, versus 4.24% for MFC-Pt. Most of the difference in soluble COD removal by MFC-EDA-1 and MFC-EDA-2 was due to methane production (0.52 g/L CH4 of 0.60 g/L total COD removal difference). The variation in COD as acetate followed similar trends to that of soluble COD although the COD as acetate measurement generally exhibited slightly higher and more stable discharge concentrations. The COD acetate analysis also yielded feed concentrations that were higher than those obtained by digestion and closer to the expected 6.44 g COD/L loading level. The reason


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 35

for this difference might be due to the presence of other soluble compounds that were detected in the IC approach at retention times lower than acetate, but could not be identified as specific compounds. It is likely that these compounds were soluble microbial products such as bacterially produced polymers, lysis products and hydrolysis products.40 Examination of the COD balances shows that most of the COD is removed by reactions that do not generate current in both MFC-EDA-1 and MFC-EDA-2 (Figure 7). MFC-EDA-1 showed a lower percentage COD removal by methane generation than MFC-EDA-2 and MFC-Pt due to its lower soluble COD. The unmeasured COD portions in MFC-EDA-1, MFC-EDA-2 and MFC-Pt were very similar and amounted to 34-39% of the total feed COD. The oxygen flux through the Nafion may have contributed significantly to the unmeasured COD removal. A rise in the amount of oxygen penetrating into the anolyte could cause biomass growth and accumulation within the MFC to increase substantially.

Figure 7

4. CONCLUSION The performance of a microbial fuel cell with a non-precious nitrogen-doped catalyst was found to be significantly better over that obtained using a conventional Pt catalyst, demonstrating promise for further application in MFCs. It is known that oxygen reduction at the cathode is a bottleneck to increasing power densities. It is reported that increasing the platinum loading from 0.1 to 0.5 mg/cm2 only increased power densities by 19%.5 This suggests that the bottleneck lies


Page 21 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

less in the catalyst used but other factors such as the active surface area for catalysis. In this work, it was found that the current/power produced varied proportionally to the loading of the non-precious metal catalyst. This provides strong evidence that the overall MFC performance was limited by the cathodic oxygen reduction reaction under the experimental conditions in this study. Results here also suggest that even higher power densities could be attained if the catalyst loading on the cathode was increased.


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1. Baffled anolyte chamber of MFC (left); constructed MFC (right).


Page 22 of 35

Page 23 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 2. (a) SEM image of catalyst (10,000x) with a portion at higher magnification (100,000x); (b) XRD diffractogram of catalyst sample.


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3. Variation of the current densities generated by MFC-EDA-1, MFC-EDA-2 and MFCPt.


Page 24 of 35

Page 25 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 4. Comparison of the polarization and power density curves generated using platinum (MFC-Pt) and non-precious metal catalyst (MFC-EDA-2).


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 5. Variation of the production rate of gases from MFC-EDA-1 (a) and MFC-EDA-2 (b).


Page 26 of 35

Page 27 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 6. Variation of the total COD content (a), soluble COD content (b) and acetate concentration (c) of the feed and discharge from MFC-EDA-1 and MFC-EDA-2.


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 7. Percentage distribution of measurable COD sinks for MFC-EDA-1, MFC-EDA-2 and MFC-Pt.


Page 28 of 35

Page 29 of 35

Energy & Fuels

Table 1. Comparison of MFC-EDAs and MFC-Pt performance 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

MFC-EDA-1

MFC-EDA-2

MFC-Pt

Average current density (mA/m ) Average power density (mW/m2) COD equivalence (mg/L-HRT)

710.7 ± 46.4 213.2 ± 13.9 339.4 ± 22.1

1439.4 ± 78.7 431.8 ± 23.6 687.5 ± 37.6

472.2 ± 8.2 141.7 ± 2.6 225.5 ± 3.9

Internal resistance (Ω)

NA

13.7

37.4

pH Eh (mV) DO (mg/L)

Feed 7.05 ± 0.06 363 ± 17 10.78 ± 1.16

Discharge 7.64 ± 0.27 87 ± 315 1.58 ± 1.24

Feed 7.01 ± 0.07 379 ± 29 10.21 ± 1.04

Discharge 7.68 ± 0.30 -2 ± 55 0.83 ± 0.73

Feed 7.03 ± 0.05 335 ± 5 15.08 ± 4.53

Discharge 7.59 ± 0.11 117 ± 78 1.63 ± 0.04

TSS (g/L) VSS (g/L)

Feed 0.38 ± 0.14 0.20 ± 0.14

Discharge 0.95 ± 0.42 0.52 ± 0.24

Feed 0.81 ± 0.87 0.25 ± 0.46

Discharge 0.67 ± 0.32 0.49 ± 0.39

Feed 0.78 ± 0.45 0.29 ± 0.02

Discharge 0.55 ± 0.14 0.20 ± 0.07

Total gas (mL/d) CH4 (mL/d) CO2 (mL/d) N2/O2 (mL/d) Methane COD Equivalent (g COD/L-HRT)

134.6 ± 8.0 82.8 ± 6.0 24.7 ± 0.6 27.1 ± 1.4 1858.3 ± 204.6

tCOD (g/L) sCOD (g/L) COD as acetate (g/L)

Feed 6.34 ± 0.74 6.36 ± 0.52 6.29 ± 0.27

CODElec (g/L) ∆sCOD (g/L) CE (%)

0.339 ± 0.022 5.06 ± 0.44 6.71 ± 0.88

0.688 ± 0.038 5.66 ± 0.69 12.18 ± 1.29

0.225 ± 0.004 6.22 ± 0.16 4.24 ± 0.41

tCOD discharge (g/L) Current generation (g/L-HRT) Methane generation (g/L-HRT) tCOD unmeasured (g/L-HRT)

1.968 0.339 1.858 2.188

0.909 0.641 2.379 2.247

0.805 0.226 2.780 2.413

2

176.8 ± 37.6 117.4 ± 27.7 44.0 ± 16.6 15.3 ± 11.6 2379.3 ± 101.4 Discharge 1.97 ± 1.15 1.30 ± 0.48 1.45 ± 0.38

Feed 6.18 ± 0.42 6.14 ± 0.60 6.64 ± 0.24


156.5 ± 26.3 130.5 ± 21.5 17.2 ± 5.0 8.8 ± 0.2 2838.9 ± 343.7 Discharge 0.91 ± 0.56 0.48 ± 0.09 0.45 ± 0.14

Feed 6.22 ± 0.29 6.22 ± 0.16 5.94 ± 0.14

Discharge 0.81 ± 0.68 0.81 ± 1.00 0.78 ± 0.22

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 35

AUTHOR INFORMATION Corresponding Author * Email: [email protected]. Tel: +1-519-4567 ext. 36728. Fax: +1-519-888-4347 Funding Sources Financial support to MP, RLL, WP and ZWC through the Discovery Grants program of the Natural Sciences and Engineering Research Council of Canada (NSERC) are gratefully acknowledged.

REFERENCES (1) Bullen, R.A.; Arnot, T.C.; Lakeman, J.B.; Walsh, F.C., Biosens. Bioelectron. 2006, 21, 20152045. (2) Bennetto, H.P.; Stirling, J.L.; Tanaka, K.; K., Vega, C.A. Biotechnol. Bioeng. 1983, 25, 559568. (3) Reimers, C. E. L.; Tender, M.; Fertig, S.; Wang, W. Environ. Sci. Technol. 2011, 35, 192195. (4) Zhang, J., PEM Fuel Cell Electrocatalysts and Catalyst Layers: Fundamentals and Applications, Springer-Verlag London Ltd., London, UK., 2008. (5) Paulus, U.A.; Schmidt, T.J.; Gasteiger, H.A.; Behm, R.J., J. Electroanal. Chem. 2001, 495, 134-145. (6) Liu, H.; Logan, B.E.; Environ. Sci. Technol. 2005, 35, 658–662. (7) HaoYu, E.; Cheng, S.; Scott, K.; Logan, B.E J. Power Sources 2007, 171, 275-281. (8) Cheng, S.; Liu, H.; Logan, B.E. Environ. Sci. Technol. 2006, 40, 364-369.

ACS Paragon Plus Environment

30

Page 31 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

(9) Deng, Q.; Li, X.; Zuo, J.; Ling, A.; Logan, B. E. J. Power Sources 2010, 195, 1130-1135. (10) Feng, L.; Chen, Y.; Chen, L.; ACS Nano 2011, 5, 9611-9618. (11) Feng, Y.; Shi, X.; Wang, X.; Lee, H.; Liu, J.; Qu, Y.; He, W.; Kumar, S.M.S.; Kim, B.H.; Ren, N. Biosens. Bioelectron. 2012, 35, 413-415. (12) Shi, X.; Feng, Y.; Wang, X.; Lee, H.; Liu, J.; Qu, Y.; He, W.; Kumar, S.M.S.; Ren, N. Bioresource Technol. 2012, 108, 89-93. (13) Zhang, B.; Wen, Z.; Ci, S.; Mao, S.; Chen, J.; He, Z. ACS Appl. Mat. Interfaces 2014, 6, 7464-7470. (14) Liew, K.B.; Daud, W.R.W.; M.; Ghasemi, J.X.; Leong, S.S.; Lim, Ismail, M. Int. J. Hydrogen Energ. 2014, 39, 4870-4883. (15) Modi, A.; Singh, S.; Verma, N., Electrochimica Acta 2016, 190, 620-627. (16) Du, H.; Bu, Y.; Shi, Y.; Zhong, Q.; Wang, J. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, 2016, 38, 527-533. (17) Ghasemi, M.; Daud, W.R.W.; Mokhtarian, N.; Mayahi, A.; Ismail, M.; Anisi, F.; Sedighi, M.; Alam, J. Int. J. Hydrogen Energ. 2013, 38, 9525-9532. (18) Zhu, N.; Chen, X.; Zhang, T.; Wu, P.; Li, P.; Wu, J. Bioresource Technol. 2011, 102, 422426. (19) Aelterman, P.; Versichele, M.; Genettello, E.; Verbeken, K.; Verstraete, W., Electrochim. Acta 2009, 54, 5754–5760 (20) Wang, L.; Liang, P.; Zhang, J.; Huang, X., Bioresour. Technol. 2011, 102, 5093–5097 (21) Xia, X.; Zhang, F.; Zhang, X.Y.; Liang, P.; Huang, X.; Logan, B.E. ACS Appl. Mater. Interfaces 2013, 5, 7862–7866 (22) Pan, Y.; Mo, X.; Li, K.; Pu, L.; Liu, D.; Yang, T., Bioresource Technol. 2016, 206, 285-289.


31

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 35

(23) Nguyen, M.-T.; Mecheri, B.; D'Epifanio, A.; Pepè Sciarria, T.; Adani, F.; Licoccia, S. Int. J. Hydrogen Energ. 2014, 39, 6462-6469. (24) Soloveichik, G. L. Chem. Rev. 2015, 115, 11533-11558. (25) Choi, J.Y.; Hsu, R.S.; Chen, Z. J. Phys. Chem. C 2010, 114, 8048–8053. (26)

Clesceri,

L.S.

Standard

Methods

for

the

Examination

of

Water

and

Wastewater; American Public Health Association: Washington, DC, 1998. (27) Liu, H.; Ramnarayanan, R.; Logan, B.E. Environ. Sci. Technol. 2004, 38, 2281-2285. (28) Beaumont, V. Investigation of Microbial Fuel Cell Performance and Microbial Community Dynamics During Acclimation and Carbon Source Pulse Tests; Master’s Thesis; University of Waterloo: Waterloo, Ontario, Canada, 2007. (29) Wu, G.; Chen, Z.; Artyushkova, K.; Garzon, F.H.; Zelenay, P. ECS Trans. 2008, 16, 159170. (30) Nallathambi, V.; Lee, J.-W.; Kumaraguru, S.P.; Wu, G.; Popov, B.N. J. Power Sources 2008, 183, 34-42. (31) Sharma, S. and Pollet, B.G.; J. Power Sources, 2012, 208, 96-119. (32) Chen, Z.; Choi, J.Y.; Wang, H.; Li, H.; Chen, Z. J. Power Sources 2011, 196, 3673-3677. (33) Zhang, J.; Wu, J.; Zhang, H., PEM Fuel Cell Testing and Diagnosis, Elsevier: Great Britain, 2013; p. 16. (34) Niwa, H.; Horiba, K.; Harada, Y.; Oshima, M.; Ikeda, T.; Terakura, K.; Ozaki, J.; Miyata, S. J. Power Sources 2009, 187, 93-97. (35) Higgins, D.C.; Chen, Z.W. Can. J. Chem. Eng. 2013, 91, 1881-1895. (36) Chen, Z.W.; Higgins, D.C.; Yu, A.; Zhang, L.; Zhang, J. Energ. & Environ. Sci. 2011, 4, 3167-3192.


32

Page 33 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

(37) Logan, B.E.; Hamelers, B.; Rozendal, R.; Schröder, U.; Keller, J.; Freguia, S.; Aelterman, P.; Verstraete, W.; Rabaey, K., Environ. Sci. Technol. 2006, 40, 5181-5192. (38) Wise, D.L. Bioprocessing and Biotreatment of Coal; CRC Press: New York, 1990; p 341. (39) Virdis, B.K.; Rabaey; Yuan, Z.; Keller, J. Wat. Res. 2008, 42, 3013-3024. (40) Laspidou, C.S.; Rittmann, B.E. Wat. Res. 2002, 36, 2711- 2720.


33

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure. Variation of the current densities generated by MFC-EDA-1, MFC-EDA-2 and MFC-Pt. 202x101mm (96 x 96 DPI)


Page 34 of 35

Page 35 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure S1. Flowchart of different stags for fabrication of the catalyst. 320x182mm (96 x 96 DPI)


Enhancement of Electricity Generation by a Microbial Fuel Cell Using

Recommend Documents