|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
Optimizing Biofuel Cell Performance Using a Targeted Mixed Me diator Combination by Jason C. Klar A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Mechanical Engineering Department of Mechanical Engineering College of Engineering University of South Florida Major Professor: Stuart Wilkinson, Ph.D. Muhammad Mustafizur Rahman, Ph.D. Frank Pyrtle, III, Ph.D. Date of Approval: March 27, 2006 Keywords: Microbial fuel cells, mediators, s. cerevisiae yeast, Neutral Red, Methylene Blue, Gastrobots Copyright 2006, Jason C. Klar
Dedication Diantha Mateos Thank you for three great years, I wi sh we had more time together You were my best friend, I love you I will carry you with me forever Salud, Amor, Dinero
iii Acknowledgements I would like to thank Dr. Wilkinson for giving me the opportunity to work on this project. Thank you for the support and for getting more out of me than I thought I had in myself. I would like to thank Shawn Applegarth for helping me with my many questions. He helped me most with formatting problems with the thesis. It was good to have someone to ask stupid questions to. I would like to thank Dr. Pyrtle and Dr. Rahman for reviewing and giving me feedback about my work. I would like to thank Sue Britten and Shir ley Tervort, in the ME department, for their support and encouragement. I would like to thank Cherine Chehab for correcting all my errors. I would like to thank Janet Giles for the extension of the deadline. And finally, I would like to tha nk my parents for their support.
i Table of Contents List of Tables iv List of Figures v List of Symbols vii Abstract ix Chapter 1 Background 1 1.1 Benefits of Biofuels 1 1.2 History of Biofuel Cells 2 1.3 Fundamentals on How Biofuel Cells Work 5 1.4 Microbial Fuel Cell Applications 6 1.5 Why Are Mediators Required? 8 1.6 The Biocatalyst 9 Chapter 2 Effects of Mediators 10 2.1 The Case for Mixed Mediators 10 2.2 The Choice of Mixed Mediators 11 2.3 Possible Mediator Interactions 13 2.3.1 Performance Using NR Only 14 2.3.2 Performance Using MB Only 16 2.3.3 Performance Using Both MB and NR 18 Chapter 3 Breakdown of Test Experiment 19 3.1 Apparatus & Methods 19
ii 3.1.1 Mechanical Components of the Fuel Cell 20 3.1.2 Physical Dimensions of the Fuel Cell 21 3.1.3 Assembly of Fuel Cell Apparatus 21 3.1.4 Wiring of the Experiment 23 3.1.5 Electrical Components 24 3.2 Sequence of Resistors 25 3.3 Fluid Systems 27 3.3.1 Gas System 27 3.3.2 Plumbing Harness 27 3.4 Bioelectrochemical Components 29 3.5 Pre-Mixing of Solutions 29 3.5.1 Buffer Solution 29 3.5.2 Substrate Solution 30 3.5.3 Catholyte Solution 30 3.5.4 Methylene Blue Mediator Solution 30 3.5.5 Neutral Red Mediator Solution 30 3.6 Adding the Solutions to the Fuel Cell 31 3.7 MFC Capping 32 3.8 Startup 32 3.9 Cleanup 33 Chapter 4 Experimental Results 34 Chapter 5 Comparison of Different Electrodes 36 Chapter 6 Yeast Staining Phenomena 38
iii Chapter 7 Hydrogen Peroxide 40 Chapter 8 Conclusion 42 References 45 Bibliography 47 Appendices 48 Appendix A: Tabulated Experimental Results 49 Appendix B: Graphical Experimental Results 52 Appendix C: Fluke Data 59
iv List of Tables Table 1: Comparison of Energy Densities for Various Sources. 1 Table 2: Structure-specific Dyes for Yeast Cells . 39 Table 3: Experimental Results for NR Only Electrodes #2. 49 Table 4: Experimental Results for NR and MB Electrodes #2. 50 Table 5: Experimental Results for MB Only Electrodes #2. 51
v List of Figures Figure 1: Electric Eel Anatomy. 2 Figure 2: Luigi Galvani. 3 Figure 3: H. Peter Bennetto. 4 Figure 4: Diagram of a MFC . 5 Figure 5: Chew Chew. 7 Figure 6: Krebs Cycle (TCA Cycle). 11 Figure 7: Matching Redox Potentials at pH 7 . 12 Figure 8: Yeast Fermentation Pathway With NR Only. 14 Figure 9: Yeast Anaerobic Respiration With MB Only. 16 Figure 10: Yeast Anaerobic Respiration With NR and MB. 18 Figure 11: Entire Apparatus. 19 Figure 12: Mechanical Com ponents of the Fuel Cell. 20 Figure 13: Microbial Fuel Cell (Gas Po rts and External Plumbing Not Shown). 21 Figure 14: Side View of Fuel Ce ll (Cathode on Left, Anode on Right). 22 Figure 15: Schematic Re presentation of Wiring of the Experiment. 23 Figure 16: Wiring of the Apparatus. 24 Figure 17: Two Resistance Decad e Boxes Wired in Series. 25 Figure 18: Selection Order of Decade Box Resistances. 26 Figure 19: Fuel Cell sh owing Plumbing Harnesses (Anode in Front, Cathode in Back). 28
vi Figure 20: Performance Results Us ing Single and Mixed Mediation. 34 Figure 21: Location Where Electrodes Were Cut from Piece of RVC Foam Plate. 36 Figure 22: Comparison of Peak Power. 44 Figure 23: Voltage vs. Resistance. 52 Figure 24: Power vs. Resistance. 53 Figure 25: Power vs. Voltage. 54 Figure 26: Log of Current vs. Voltage. 55 Figure 27: Power vs. Voltage for Different Electrodes. 56 Figure 28: Power vs. Resistan ce for Different Electrodes. 57 Figure 29: Comparison of Hydrogen Peroxide to Ferricyanide in the Cathode. 58 Figure 30: Fluke Data for No Mediator. 59 Figure 31: Fluke Data for MB. 60 Figure 32: Fluke Data for MB and NR. 61 Figure 33: Fluke Data for NR. 62 Figure 34: Fluke Data for MB With H2O2 in the Cathode. 63
vii List of Symbols ADH alcohol dehydrogenase ADP adenosine diphosphate ATP adenosine triphosphate CoQ coenzyme Q, ubiquinone DAPI 4,6-diamidino-2-phenylindole DARPA Defense advanced research project agency DNA deoxyribonucleic acid e.coli Escherichia coli E o redox potential FAD flavine adenine dinucleotide FADH2 reduced form of flavine adenine dinucleotide GAPDH glyceraldehydes-3phosphate dehydrogenase GTP guanosine triphosphate H+ hydrogen ion H2O2 hydrogen peroxide IDH isocitrate dehydrogenase KDH -ketoglutarate dehydrogenase K3Fe(CN)6 potassium ferricyanide MB methylene blue MBH2 reduced form of methylene blue
viii MDH L-malate dehydrogenase MFC microbial fuel cell NAD nicotinamide adenine dinucleotide NADH reduced form of nicotinamide adenine dinucleotide NADP nicotinamide adenine dinucleotide phosphate NR neutral red NRH2 reduced form of neutral red OCV open circuit voltage PDH pyruvate dehydrogenase RVC reticulated vitreous carbon SDH succinate dehydrogenase SHE saturated hydrogen electrode SUC/FUM succinate/fumerate system TH thionine
ix Optimizing Biofuel Cell Performance Using a Targeted Mixed Me diator Combination Jason C. Klar ABSTRACT A study of how mediators in teract with the catabolic pathways of microbes was undertaken with a view towards improving the performance of microbial fuel cells. The use of mediators is known to improve the power density in microbial fuel cells, but this work suggests that no single mediator is idea lly suited to the task. Instead, a carefully selected mixture of two targeted mediators (Methylene Blue and Neutral Red) might be optimal. To test this hypothe sis, a yeast-catalyzed microb ial fuel cell was built and empirically evaluated under different medi ation conditions while keeping all other parameters constant. The results clearly show that an appropr iate mix of the two mediators mentioned could inde ed achieve significantly supe rior performance, in terms of power-density, than when either mediator is used singly. All tests were carried out using the same overall mediator concentration.
1 Chapter 1 Background 1.1 Benefits of Biofuels Biomass is very diverse, ranging from gr asses and grains to animal products such as meat or fish. As can be seen from Tabl e 1, the energy density of biomass is generally inferior to fossil fuels, but it can be cataboliz ed into electricity rather than combusted, resulting in a better overall efficiency compar ed to that of current power plants that use fossil fuels. The energy density of biomass is vastly superior to leading chemical batteries, offering the promise of extended operation fo r portable electronic devices such as cell phones, laptop computers, palm-pilots, and GPS systems. EnergyEnergy DensityDensity Energy Sources Type [kcal/g] Energy Source Type [kcal/g] Hydrogen Gas 33.921 Gasoline Octane 11.490 Carbohydrate D-Glucose 4.061 Animal Fat Saturated 9.076 Coal Coke 7.000 Vegetables Above Ground 0.241 Beef (Lean) 2.750 Soft & Juicy 0.616 Meat Chicken 1.500 Fruit Citrus 0.372 Termites 3.500 Li/MnO2 0.258 Insects Grasshoppers 2.000 Primary Batteries Zn/MnO2 0.112 Waxworms 1.811 Ni-MH 0.060 Worms Mealworms 1.223 Secondary Batteries Ni-Cd 0.039 Table 1: Comparison of Ener gy Densities for Various Sources.
2 1.2 History of Biofuel Cells As with all branches of science, wh at we know today about biofuel cells represents the accumulated effort of many. Incredible discoveries and innovations continue to be made today. Ho wever, it all had to start with the pioneering work of just a few. People have known for millennia that some animals have the ability to create electricity. Hieroglyphics from Horapollo dated around 3000 BC depict this ability in Electric Catfish in the Nile, wh ich release a charge as a defense mechanism. Studies of electric fish anatomy eventually led to the discovery that acetylcholine is a vital ingredient to the electrochemical transmi ssion of nerve impulses. Acetylcholine is broken down by the enzyme cholinesterase and all animals have both these biochemicals. Electric eels have a much higher (1000 times ) quantity of cholinesterase, which is probably why they can release electric charges up to 600 volts strong . Fi g ure 1: Electric Eel Anatom y Vital organs located in front 20% of body Snake-like body contains electric-generating organ Gill cove r Small e y es Elon g ated anal fin Skin has tin y scales
3 Figure 2: Luigi Galvani. Luigi Galvani, in 1791, was the first person to show how electricity could be related to biological organisms. He demonstrated this basic link when he applied a voltage to the legs of a frog and observed a muscle spasm in response. Galvani developed a theory of animal electricity, that all animals have an electrical ne rve fluid that reacted to a completed electrical circuit and caused the muscles of even a dead frog to contract. Galvanis theory was not accepted until Carlo Matteucci was able to prove it in 1831. Matteucci showed, when injured, the frog emitted a small amount of electrical current using a sensitive galvanometer. This injury current didnt require the aid of metallic or atmospheric electricity. He was not able to detect it in the nervous system, only from the wound itself. Michael Cresse Potter, in 1910, demonstr ated that organisms (and enzyme extracts) could generate voltage and deliver current. He placed one Platinum electrode in an anaerobic culture containing glucose and yeast (or Escherichia coli ), and a second in a blank aerobic culture containing no microbe s. He recognized intuitively that the electrons came from the degr adation of food in the organi sms, but little was known about biochemistry or metabolic processes at that time. This is the earliest documented example of a biofuel cell. In 1931, Barnett Cohen described how a ba ttery of such biofuel cells could produce more than 35 volts. Cohen also used benzoquinone or ferricyanide in the anode
4 Figure 3: H. Peter Bennetto. compartment, the first recorded example of the use of chemical mediators to aid in electron transport. In 1963, Milton J. Allens pioneering studies of bacterial electrophysiology at a number of US institutions, were designed to elucidate the metabolic behavior of E.coli These studies eventually lead to the discover y of the respiratory metabolic apparatus of living organisms (TCA cycl e) as shown in Figure 6. The catabolic generation of electricity from methane and higher hydrocarbons was soon pioneered, along with the earliest enzyme-based and photo-microbial fuel cells. Significant contributions were made by: John Davis & Henry Yarborough of the Mobile Co. used Nocardia (1962) also used enzymes William van Hess used Pseudomonas methanica with CH4 fuel (1964) Hector Videla used Micrococcus certificans in Argentina (1972) Yahiro used Glucose Oxidase in an enzymatic fuel cell (1964) Berk & Cransfield used photosynthetic Rhodospirillum rubrum in a photomicrobial fuel cell (1964) In 1980, H. Peter Bennetto and his bioelectrochemistry group at Kings College, London, contributed greatly in the area of redox mediators as applied to Microbial Fuel Cells (MFCs). The group included: Thurston, Roller, Stirling, Delaney, Mason, and Tanaka (visiting from Japan).
5 Many fuels have been used in biofuel cells, including: sugars, alcohols, urea, hydrocarbons, sulfide; and ev en natural biomass such as : molasses, coconut oil, cornhusks, milk whey, fishmeal, plankton, etc. Research in MFCs is an ever-expanding field. 1.3 Fundamentals on How Biofuel Cells Work The biofuel cell, otherwise known as MFC has been demonstrated  as a device that is capable of efficiently converting va rious food substrates such as carbohydrates, sugars, fats, etc. directly into electricit y without combustion, using microorganisms as biocatalysts. The MFCs basic set up consis ts of an anode and a cathode separated by a proton exchange membrane. The anode cont ains a biocatalyst (microbe) along with a substrate (sugar) in a buffer solution. As th e microbe breaks down the substrate, ions (H+) and electrons (e-) are made available within the anode chamber along with some Figure 4: Diagram of a MFC .
6 CO2. In normal aerobic respiration, the ions and electrons would combine with oxygen to create H2O. However, this cannot occur due to the anoxic conditions maintained within the anode chamber. Instead, the cathode chambe r is enriched with oxygen or some other electron accepter. The cell, in balancing the electric charge allows the ions to pass through the proton exchange membrane exclusiv ely, while electrons must travel to the cathode via an external circuit, and hence do useful work on a load. Once on the cathode side of the cell; the ions, electrons and oxygen molecules combine to produce H2O. 1.4 Microbial Fuel Cell Applications Popular portable consumer electronic de vices need light-weight, small, extended operation power sources to replace current battery technology. Due to limited power density, biofuel cells are most likely to appear first as portable batte ry chargers rather than as direct ba ttery replacements. Medis Technologies Ltd is currently working on an ethanol-fueled charger for the U.S. Arm y. When combined with MEMS or nanotechnologies, biofuel cells may one day b ecome viable no-maintenance, extended-use battery replacements at ambient operating te mperatures. The long recharge times associated with batteries would be elimin ated by the quick addition of fresh fuel cartridges. Such a lucrative market ha s attracted a large num ber of development companies: Powerzyme Medis Technologies MesoFuel Manhattan Scientifics Lilliputian Systems Angstrom Power and MTI MicroFuel Cells Most utilize alcohol fuel, but not all are adopting a bio-catalyzed approach; some use noble metals, extreme pH, and elevated temperatures.
7 Tiny enzymatic fuel cells are under development which can be surgically implanted into blood vessels, and which utilize blood sugar as fuel. Potential applications include power sources for implantable devices, such as: tracking devices medical sensors telemetry chips pacemakers data storage Another potential application for biofuel cells is in Gastrobots or food-powered robots, which are autonomous, self-sufficien t foraging machines. Their food source can potentially be any biological food. Such sy stems represent an ideal biomimetic solution to energy demand during long-term start-and-forget missions. Gastrobots represent an immense technological challenge, since they must find and ingest complex biomass while tolerating no intervention. Developed at USF in 2001, by Dr. Stuart Wilkinson, Gastronome (a.k.a. ChewChew) , the worlds first food powered robot was built to utilize sugar only as seen on Figure 5. Such a diet has clear advantag es for a pioneering pr ototype, but is not applicable to a self-sufficien t gastrobot sustained through fo raging . Other gastrobot applications include long-range underwater systems. These aquatic robots would be able to travel great distances, only surfacing to lo cate position and to deliver data via satellite. The food source for such a device is likely to be plankton or small fish. DARPA is currently funding such a system.
8 1.5 Why Are Mediators Required? It is well known that mediators can dr amatically improve the performance of whole-cell biofuel cells (or Microbial Fuel Cells MFCs) by acting as an electron shuttle between intracellular reducing centers and an external electrode . Many different mediators have been experi mentally studied [6, 7], but relatively little has been reported on wh y certain mediators work be tter than others [8, 9]. Mediators are generally used singly, such th at only a few examples  exist in the literature where combinations of mixed me diators have been em ployed. The present study represents an early attemp t to tailor a mixed mediator combination for a specific biocatalyst organism, with the goal of im proving MFC performance significantly beyond single mediator levels. Figure 5: Chew Chew.
9 1.6 The Biocatalyst Since mediators fundamentally work by in teracting with the metabolic pathway of the biocatalyst, it was advantageous to choose a well-studied microorganism. Top candidates were e.coli and yeast ( saccharomyces cerevisiae ) since both are facultative anaerobic organisms able to switch metabolism from respiratio n in the presence of oxygen to fermentatio n when anoxic. However, with yeast the metabolic pathway during fermentation is extremely simple, resulting in primary end products of ethanol and CO2. Yeast was therefore chosen as the biocatalys t since the organisms simple fermentation minimized the number of potential mediator inte raction sites, and cons equently helped to elucidate the fundamental pro cesses involved. Yeast has been successfully used in MFCs in the past [11, 12]. An additional benefit of using s.cerevisiae is that it works well at room temperatures (20 25 C), unlike e.coli that prefers the elevated temperatures associated with its favored enteric habitat. S.cerevisiae is also safe and very easy to handle, while being readily available from brewing suppliers.
10 Chapter 2 Effects of Mediators 2.1 The Case for Mixed Mediators Mediators are fundamentally lipophilic chemi cal electron carriers that are able to pass through the cell walls of microorganisms, and thereby move between intracellular space and the extracellular environment. They are able to assist the microorganisms metabolism in an anoxic environment by acting as a terminal electron acceptor. Once inside the microbial cell, they are believed to interact with the metabolic process by participating in redox reactions with bioelectrochemical subs tances such as the pyridine nucleotides (NAD, NADP), flavoproteins iron-sulfur prot eins, quinones, and cytochromes. Conventional wisdom would indicate the us e of a mediator with a low (negative) redox potential, and preferably one with a formal potential clos e to that of the pyridine nucleotides, as this will maximize the open circuit voltage (OCV) of the MFC and facilitate efficient transfers . However, such mediators may not necessarily generate high power-density because th e microbe is generally biased towards a fermentative metabolism due to the absence of oxygen, or an y other suitable terminal electron acceptor of sufficiently positive potent ial in the MFC anode chamber. Certain mediators are, however, able to act as terminal electron acceptors in the absence of oxygen, and in so doing enable a facultative anaerobe to switch to a nearly
11 Figure 6: Krebs Cycle (TCA Cycle). GTP: guanosine triphosphate FAD: flavine adenine dinucleotide N AD: nicotinamide adenine dinucleotidecomplete respiratory metabolism ( anaerobic respiration ), rather than the fermentation normally demanded by such an anoxic en vironment. This benefits the MFC performance by enabling more complete oxidat ion of the fuel, and by generating a larger pool of reducing equivalents. However, to be effective as a terminal electron acceptor the mediator requires a somewhat positive redox pot ential, preferably one that is close to that of a carrier already in th e respiratory chain. By virtue of its positive potential this type of mediator is less able to ta p the pool of reduced intermediates. Clearly the above arguments are mutually exclusive for any single mediator. Instead a mixture of two mediators is ca lled for, one of somewhat positive redox potential to trigger a switc h towards anaerobic respirat ion, combined with one of substantially negative redox potential to expl oit the pool of reduced pyridine nucleotides (NADH) present. 2.2 The Choice of Mixed Mediators In yeast, as with most othe r microbes, it is the pyridine nucleotides that constitute the bulk of the reducing equivalents generate d. A yeast-catalyzed MFC would therefore
12 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 pH Redox Potential, Eo (vs. SHE) [V] FerriFerrocyanide MB SUC/FUM N R N AD / NADH 7 311 Figure 7: Matching Redox Potentials at pH 7 . benefit from the presence of one mediator with a redox potential very close to that of NAD (i.e. 0.32 V) to faci litate the direct and efficient transfer of redu cing-power from this cofactor. As for selecting the second mediator, it is clear that the TCA cycle, see Figure 6, cannot become fully established in an an aerobic environment without an electron acceptor of sufficiently positive redox potential. This relates to the step in the cycle where succinate is oxidized to fumarate by succinate dehydrogenase (SDH). The redox potential of the succinate/fumerate (SUC/FUM) system is +0.03 V, which makes it a very weak and unsuitable reductant for NAD or, presumably, any mediator of similarly negative redox potential. Instead, the TCA cycle uses protein-bound FAD as the hydrogen carrier for the SUC/FUM system. Normally the FADH2 generated in the SUC/FUM system would feed hydrogen into the respiratory chain via coenzyme Q (CoQ), but in the absence of oxygen this chain leads to a dead-end, thereby precluding the entire respir atory apparatus. However, by providing a mediator with a redox potential close to that of the SUC/FUM system, the FADH2 can become reoxidized and the hydrogen transporte d out of the microbe via this terminal electron acceptor. For mediat or reduction to occur preferentially to CoQ, and thereby
13 maximize OCV, the former should possess a less positive redox potential than the latter (< +0.113 V). In summary, it would appear that supe rior performance would result from a yeast-catalyzed MFC if dual mediators are us ed; one with a redox potential close to 0.32 V and the other with a redox potential close to +0.03 V but less positive than + 0.113 V. Redox potentials are a function of pH, but given that the MFC is to operate at physiological conditions it was only necessary to find two mediators that possessed the necessary potentials at pH 7. Figure 7 shows how Neutral Red (NR) at Eo = -0.325 V clearly represents a ne ar perfect match for NAD at pH 7. Indeed NR has been shown to chemically reduce NAD in vitro . Methylene Blue (MB) at Eo = +0.011 V is a good choice for the second mediator, as complete equilibrium between MB and SUC/FUM has been demonstrated in classical studies [16, 17]. Thionine (TH) at Eo = +0.064 V may be a satisfactory alternate to MB , albeit a more expensive one. 2.3 Possible Mediator Interactions The interactions represented in the followi ng are rather simplistic in that they are based solely on redox potential and the catabol ic pathways. The effect of chemical kinetics and polarizations are not considered, nor are interactions that might occur during anabolism. None-the-less these speculati ons do correspond generally to the broad outcomes observed during experimentation. More detailed experiments are still needed to confirm or deny the hypothesis presented in the following. How deleterious mediators are to living cells is not clea r, but at the very least, robb ing a cell of reducing power that would normally be employed in ATP synthe sis can only be considered antagonistic.
14 2.3.1 Performance Using NR Only Normal yeast fermentation of glucose us es NAD as the sole electron carrier, acting as a reducing coenzyme for gl yceraldehyde-3-phosphate dehydrogenase (GAPDH) and as an oxidizing coenzyme for alcohol dehydroge nase (ADH) in balanced reactions requiring no external hydrogen accepto r. Two NADH are thus generated for each mole of glucose fermented. The additi on of NR is unlikely to precipitate any switch to anaerobic respiration (given its lo w redox potential), and so it will most likely interact with the fermentative pathway. Figur e 8 shows how this interaction could occur in three possible ways, denoted by n = 0, 1 or 2. It is quite possible that at any given moment or locati on in the MFC, no oxidized NR has permeated inside a particular yeast cell. This scenario is indicated by n = 0, and fermentation will proceed as normal with th e production of two mol of ethanol and no contribution to the MFC output. The case where n = 1 suggests that some intracellular oxidized NR is available and consequently the reduced NADH is partly reoxidized by ADH and partly by the NR. This latter case will only generate one mol of ethanol per mol of glucose, leaving one mol of acetal dehyde unoxidized and providing a limited n(NAD)n(NR)m(Ethanol) GlucoseANODEn = 0, 1 or 2 m = 2 n 2(Acetaldehyde) 2(Pyruvate) GAPDH ADH 2CO2 m(NAD) -0.32V redredred ox ox ox 2ATP 2ADP Figure 8: Yeast Fermentation Pathway With NR Only.
15 flow of reducing power via the anode. Fi nally, when n = 2 no ethanol is produced and the entire reducing power of 2 mol NAD per mo le of glucose is exchanged with the anode. This scenario is most likely to occur under conditions of high mediator concentration where the yeast ce lls are saturated in NR and th e chemical kinetics are not seriously rate limiting. The greatest MFC output is anticipated under these circumstances, but it is unclear whether a corresponding buildup of acetaldehyde will occur or if the yeast will terminat e its metabolic pathway at pyruvate. The anodic half reaction for n = 2 becomes: Glucose + 2ADP + 2NR 2Acetaldehyde + 2ATP + 2NRH2 + 2CO2 The free energy is expressed by Go = eF Eo, where eF = 46 kcal/V for the transfer of two electrons (e = 2). If an oxygen cathode (Eo = +0.82 V) is employed in the MFC the 2NRH2 represents a free energy of 2 [46 (0.82 + 0.32)] = 105 kcal per mol of glucose or 15% energy efficiency (given that the free energy of oxidation for glucose is 686 kcal/mol).
16 2.3.2 Performance Using MB Only In the presence of a suitable terminal hydrogen acceptor such as MB, yeast is likely to switch to anaerobic respiration. This initially involves the Embden-Meyerhof pathway from glucose to pyruvate, which includ es one NAD reduction. Pyruvate is then oxidized to acetyl-CoA by the PDH complex with one more NAD reduction, after which the TCA cycle adds a further three NAD reducti ons. These latter three are associated with the -ketoglutarate dehydrogenase (KDH) complex, L-malate dehydrogenase (MDH) and NAD-specific isocitrate dehydrogena se (IDH). With two mol of pyruvate generated and two turns of the TCA cycle needed to completely oxidize a mole of glucose, ten NADH are gene rated per mole of gluc ose by respiring yeast. As depicted in Figure 9, the primary ro le of MB is as a terminal hydrogen acceptor ideally matched to the redox potenti al of the SUC/FUM system. MB thereby acts as a surrogate for coenzyme Q and enab les a switch to anaerob ic respiration. Yeast TCA CycleNAD NAD NAD NAD NAD FAD MB ANODE SDH Acetyl-CoA1/2 Glucose E-M-P Glycolysis IDH MDH KDH Pyruvate ox red MB ox red MB red MB ox red MB ox red MB ox red SUC/FUM System PDH -0.32V +0.01V 15kcal FMN Fe-S ox FMN Fe-S FMN Fe-S FMN Fe-S FMN Fe-S ATP ADP ADP ATP ADP ATP oxox ox ox ox red red red red red ox Fi g ure 9: Yeast Anaerobic Res p iration With MB Onl y .
17 respiration clearly offers five times the NAD H yield as compared with fermentation, but since MB is only a weak reductant for NA DH it is not well suited to tapping these reducing equivalents directly. The yeast must employ the flavoproteins (FMN) and ironsulphur proteins (Fe-S) of its normal resp iratory chain to affect the reoxidation of NADH. The aforementioned proteins allow the coupling of NAD/NADH at -0.32 V to MB/MBH2 at +0.011 V via a series of reducing steps that corres ponds to a free energy of 15 kcal/mol of NAD. Given that ten NADH ar e produced per mol of glucose, this NAD/MB coupling apparatus represents an MFC output energy loss of 150 kcal/mol of glucose. The yeast may, however, benefit sinc e it is able to generate ATP from this coupling site in the same way as it does when coupling NAD to CoQ in normal aerobic respiration. The anodic half reaction for th e case of MB only becomes: Glucose + 6H2O + 14ADP + 12MB 14ATP + 12MBH2 + 6CO2 In this case 2ATP is generated during gl ycolysis, 2ATP by the TCA cycle itself, and 10ATP by electron-trans port phosphorylation. When employing an oxygen cathode in the MFC, the 12MBH2 represents a free energy of 12 [46 (0.82 0.011)] = 447 kcal per mol of glucose or 65% energy efficiency.
18 2.3.3 Performance Using Both MB and NR As can be seen in Figure 10, the effect of including both MB and NR is to eliminate the need for the FM N and Fe-S, as NR and NAD ar e able to couple directly. This deprives the yeast of 10ATP but makes this extra energy available to the MFC via the oxidation of NR at the anode. The anodic half reaction for mixed MB and NR becomes: Glucose + 6H2O + 4ADP + 2MB + 10NR 4ATP + 2MBH2 + 10NRH2 + 6CO2 In this case 2ATP is generated during glycolysis and 2ATP by the TCA cycle itself. When employing an oxygen cathode in the MFC, the 2MBH2 + 10NRH2 represents a free energy of 2 [46 (0.82 0.011)] + 10 [46 (0.82 + 0.32)] = 599 kcal per mol of glucose or 87% energy efficiency. TCA CycleNAD NAD NAD NAD NAD FAD NR SDH Acetyl-CoA 1/2 Glucose E-M-P Glycolysis IDH MDH KDH Pyruvate No ATP from Respiratory Chain ox red NR red MB red NR ox red NR ox red NR ox red SUC/FUM System PDH -0.32V ox ox ANODE ADP ATP ATP ADP ox oxox ox ox red red red red red ox red Fi g ure 10: Yeast Anaerobic Respiration With NR and MB.
19 Chapter 3 Breakdown of Test Experiment 3.1 Apparatus & Methods A small desktop apparatus was built to te st the hypothesis that a yeast-catalyzed MFC will perform better under mixed mediation than when mediators are used singly at the same overall mediator concentration. Apar t from mediation, all ot her test parameters were kept constant throughout the main experimental program. Figure 11: Entire Apparatus.
20 3.1.1 Mechanical Components of the Fuel Cell Compartments: Annular virgin-grade PTFE (Teflon) End Plates: Square transparent Plexiglas Seals: Silicon rubber o-rings Electrodes: Reticulated vitreous carbon (RVC), 100 pores per inch Membrane: Nafion 115 Current Collector: Square carbon felt washer Hardware: All Stainles s Steel bolts and nuts Figure 12: Mechanical Components of the Fuel Cell. DESKTOP APPARATUS Components RVC Electrode Nafion Membrane Teflon Cell Rin g Silicon O-Ring Plexiglas End Plate Carbon Felt Washe r Stainless Steel Hardware Fluid Inlet/Outlet ( w/O-rin g seals ) Terminal (w/O-ring seal)
21 Plexiglas End Plate Screw Terminal Nafion Membrane O-Ring Seal Teflon Chamber RVC Electrode Figure 13: Microbial Fuel Cell (Gas Ports and External Plumbing Not Shown). 3.1.2 Physical Dimensions of the Fuel Cell Compartment Volumes: 32 mL each (both anode and cathode). Electrodes: Circular, 4.22 cm diameter x 0.584 cm thick. Electrode Geometric Area: 30 cm2 of exposed RVC (as used in power-density calculations). Membrane Exposed Area: Circular, 5 cm diameter. 3.1.3 Assembly of Fuel Cell Apparatus The chambers for the anode and cathode were contained by the annular virgingrade PTFE (Teflon) with a Nafion memb rane sandwiched between them. Leakage was prevented between chambers and membrane using Silicon rubber o-rings. A small amount of silicon grease was used to hold the o-rings in place dur ing assembly. RVC was chosen for the electrodes due to its c onductive properties, its biocompatibility, and
22 high surface area to volume ratio. Stainless steel hardware that was attached to the electrodes protruded through the endplates, thereby making electr ical screw terminals. A carbon felt washer was used on bo th electrodes to cushion th is friable material during clamping and to act as current collectors. Each electrode was positioned within the chamber equidistant from the NafionTM membrane and the PlexiglasTM end plate. A different membrane was used for each of the four mediation conditions of: no mediator, Methylene Blue, Neutral Red, an d mixed Methylene Blue with Neutral Red. This step ensured that there would not be any MB cont aminating the NR test or vice versa. The endplates were square transparent Plexiglas to create a stable base and to allow observation of the internal bioelectrochemi cal components. The entire apparatus was held together and easily disassemble d using four bolts with wing nuts. Figure 14: Side View of Fuel Cell (Cathode on Left, Anode on Right).
23 3.1.4 Wiring of the Experiment The MFCs terminals were connected dire ctly across two resi stance decade boxes in series. The voltage across these resi stance decade boxes was monitored using a Fluke Data Bucket 2635A precision multimet er and logger. The logger recorded voltage at precisely five-second intervals. Data was also se nt to a P.C. running Fluke Hydra Logger and Trend Link software to further monitor the voltage. The wire used in the apparatus was 12-AWG copper. A knife switch was include d in the circuit to switch between open-circuit and loaded modes of operation. Figure 15: Schematic Representation of Wiring of the Experiment. Resistance Decade Boxes Fuel Cell FlukeBox To Data Logger and PC Knife Switch
24 3.1.5 Electrical Components Real-time Monitoring: P.C. running Fluke Hydra Logger and Trend Link software. Circuit Control: Knife switch to select between open-circuit or loaded condition. Electrical Measurements: Fluke Data Bucket 2635A precision multimeter and logger. Electrical Loads: Resistance decade boxes using 53 values ranging from 1,000,080 down to10 Figure 16: Wiring of the Apparatus.
25 Figure 17: Two Resistance Decade Boxes Wired in Series. 3.2 Sequence of Resistors Output voltage data from the MFC was coll ected at different loads. These loads were provided by two electrical resistance deca de boxes wired in series. Each decade box had five selector switches, which were used to adjust the total resistance of the box. All five of these switches were used in the case of the first box, while only one switch was utilized on the second box. At the outset of the test, the first box had all its switches set to their maximum values (marked 9 in each case). This resulted in a total resistance for the first box of 999,990 (i.e. 9 x 105 + 9 x 104 + 9 x 103 + 9 x 102 + 9 x 101 ). The second box used only one sw itch, which was se t to just 90 (i.e. 9 x 101 ). The remaining four switches on the second box were set to 0 or short circuit. Since the two boxes were wired in series the initial resistan ce across the MFC was 1,000,080 (i.e. 999,990 + 90 ). During MFC testing the load resistance was incrementally reduced by switching just one switch at a time, the sequence was as follows: The resistance decade box was firs t started at a maximum value of 1,000,080 then proceeded to drop at 100,000 increments until it reached 100,080 see curve 1 on Figure 18. At 100,080 it dropped at 10,000 increments until it reached 10,080 see curve 2 on Figure 18. At 10,080 it dropped at 1,000 increments until it
26 reached 1,080 see curve 3 on Figure 18. At 1,080 it dropped at 100 increments until it reached 180 see curve 4 on Figure 18. At 180 it dropped 10 increments until it finally reached 10 see curve 5 in Figure 18. The reason 90 was added using a second resistance decade box was because the slope of the power-resistance curve changed rapidly so more data points needed to be obtained between 100 and 200 to give a smoother repres entation of the graph, see Appendix 9. Load resistance was changed consistently using the decade boxes and voltage was recorded once the value stabilized each time Once resistance and vol tage data had been collected, current and power coul d be calculated and tabulated. 1 10 100 1,000 10,000 100,000 1,000,000 10,000,000 1357911131517192123252729313335373941434547495153Recorded Interval [unit-less]Logaritum of Load Resistance [log ] Figure 18: Selection Order of Decade Box Resistances. 1 2 3 4 5
27 3.3 Fluid Systems 3.3.1 Gas Systems An air pump was used to percolate air th rough the aquarium air-line tubing which was attached to the cathode to stimulate mixing and to introduce more oxygen in the catholyte. The anolyte was agitated using a constant flow of 99.9999% N2 using nonpermeable Tygon tubing and thereby purged of O2. A flow meter was employed to maintain a constant flow of the Nitrogen; how ever, different flow rates were explored to maximize the fuel cells effectiveness. It wa s noticed that too much flow of Nitrogen was detrimental to the performance of the MF C, presumably because the gas pockets prevented adequate contact between electroa ctive components and th e electrode surface. 3.3.2 Plumbing Harnesses The anodes plumbing harness included a series of check valves, Tygon tubing, tubing tees, and a syringe. Pl unging of the syringe would st ir the contents of the MFC and give the anolyte more uniformity. It wa s noted that mixing of the anode using this method after the fuel cell was running was detr imental to the performance of the MFC, presumably because it dislodged the microbes (yeast) from their locations on the electrode surface and thereby hindered electr on exchange. The syringe plunging system was invaluable in drawing the anolyte compone nts into the chamber in itially and was also beneficial in the cleanup proc ess by evacuating most of the liquid before having to unbolt it. Three entire anode harnesses were create d, one for each: no mediator which was later
28 converted into NR only, MB only, and mixed MB with NR. This was a precautionary measure to eliminate contamination between the tests. The cathodes air harness differed from the anodes because it us ed aquarium airline tubing instead of the expensive Tygon tubing. The cathode had atmospheric air pumped into it, so the non-permeable tubing was not necessary. Th e cathode also had a syringe and a check valve system that was utilized during filling, mixing and cleanup processes. Figure 19: Fuel Cell Showing Plumbing Harnesses (Anode in front, Cathode in back). Air pump Catholyte Syringe Header tubes Check Valve Nitrogen inlet tube Anolyte Syringe
29 3.4 Bioelectrochemical Components Buffer: 0.1 M Phosphate (mixed potassium & sodium) @ pH 7. Substrate (final concentration in anolyte): 50 mM dextrose. Biocatalyst: 5/8 g dry wt (25 mg/mL of anolyte, 1010-1011 cells) S. cereivisiae (Red Star Pasteur Champagne, active dry wine yeast, Lesaffre Corp). Mediators: Methylene Bl ue (Riedel-deHaen 32723 > 98%), Neutral Red (SIGMA N7005 > 96%). Mediation (final concentration in anolyte): 1 mM. Anolyte: Substrate + Biocatalyst + Me diator, in 0.1 M Phosphate Buffer, pH 7. Catholyte: 0.1 M Potassium Ferricyanide, K 3 Fe(CN) 6 in 0.1 M Phosphate Buffer, pH 7. 3.5 Pre-Mixing of Solutions 3.5.1 Buffer Solution A 0.1M Phosphate buffer at pH 7 was cr eated using 200mL distilled water and a combination of 1.0534 grams of Potassi um Phosphate (Monobasic) and 1.74 grams Sodium Phosphate in precise amounts. Then the pH of the buffer was tested using a digital pH meter. Final minor adjustments were then made as follows: if the pH was too acidic, Sodium Phosphate was added and if the pH was too basic, Potassium Phosphate was added to get a desirable pH of 7 0.01. This buffer wa s used to prepare both the anolyte and catholyte.
30 3.5.2 Substrate Solution 20 mL of sugar solution was prepared pr ior to each test, although only 17 mL was actually added to the anode chamber to achieve the correct concentration. To prepare the sugar solution 0.339 grams of dextrose was adde d to 20mL of buffer solution to create a concentration of 94 mM. 3.5.3 Catholyte Solution A 0.1M Potassium Ferricyanide soluti on was prepared by adding 1.054 grams of the reddish-orange crystals to 32 mL of buffe r solution. Swirling was required to ensure that the Ferricyande had completely dissolved. 3.5.4 Methylene Blue Mediator Solution Methylene Blue (MB) has a formula we ight of 319.86 g/mol. A batch of this mediator solution was prepared by adding 0.0512 grams of MB to 50 mL of buffer to create a concentration of 3.2 mM. 3.5.5 Neutral Red Mediator Solution Neutral Red (NR) has a formula weight of 288.78 g/mol. A batch of this mediator solution was prepared by adding 0.0462 grams of NR to 50 mL of buffer to create a concentration of 3.2 mM.
31 Each of these solutions were swirled until all solid material had fully dissolved. Excessive agitation was avoided so as not to oxygenate these solutions. Once all solutions had been prepared, testing was commenced as soon as possible because the yeast will become less effective over time. 3.6 Adding the Solutions to the Fuel Cell Prior to collecting data, the MFC was pr epared by adding ingredient solutions. 1. 10 mL of 0.1 M Ferricyanide was added into the top of the cathode using a syringe. Different syringes were used fo r dispensing each ingr edient so as to avoid cross contamination, especially si nce ferricyanide is toxic to yeast. 2. 10 mL of 94 mM sugar solution was adde d into the top of the anode using a second syringe. 3. 10 mL more of 0.1 M Ferricyanide was adde d into the top of the cathode using the cathode syringe. 4. 7 mL of 94 mM sugar solution was added into the anode using the anode syringe, completing the desired amount of sugar solution. 5. 12 mL of 0.1 M Ferricyanide was added in to the top of the cathode using the cathode syringe, completely filling the cathode. 6. 5 mL of yeast was added into the anode, using the anode syringe. 7. 10 mL of 3.2 mM mediated buffer was added into the anode creating a final concentration of 1.0 mM mediated buffer, completely filling the anode. This was
32 the case when using MB or NR exclusiv ely. For the case of mixed mediators, 5 mL of MB together with 5 mL of NR were both added. Filling the chambers slowly and evenly was important in eliminating bending of the membrane, which, in severe cases can lead to shorting of the electrodes. Since 32 mL of the various solutions were added to bot h the anode and the cathode chambers, their height levels were the same. This was ch ecked visually through th e plexiglas endplates to confirm a successful fill procedure. At this point, all solutions have been added to the fuel cell. 3.7 MFC Capping Having added all the solutions to the MF C, the two chambers were each capped off using a small header tank (made from syri nge bodies). They were held in place by inserting their nozzle ends into a small hol e on top of the anode and cathode chambers. These holes were needed to allow gases to escape. A cotton ball was placed on the top of each header tank to catch any anolyte/catholyt e that splashed up through the small holes due to the force of escaping gases. (N2 for the anode, pumped air for the cathode) 3.8 Startup To start the experiment, the Fluke Hydra Logger and Trend Link programs were activated. Once started, values for vol tage were witnessed on the P.C. monitor using the Hydra Logger program, and also di rectly displayed on the digital output for
33 the Fluke Data Bucket 2635A Multimeter. Th e air pump was then turn ed on to aerate the cathode and to maintain an elevated oxygen mi xture. The Nitrogen tank was then turned on to agitate the anolyte and purge oxygen. Flow was adjusted to a constant flow rate. The voltage would then climb until it reached a peak; this peak value represented the Open Circuit Voltage (OCV). Starting at the largest resistance, the load was reduced one increment at a time, while recording voltage at each increment. This procedure ensured only light current draw initially ; heavy current draws tend to de plete the charge within the cell faster than the microbes can maintain it. 3.9 Cleanup After data was collected, it was important to thoroughly clean the apparatus prior to commencing the next test. Th is involved the following steps: Electrode Preparation: Washed in 70% ethyl alcohol, then rinsed in distilled water, then washed in muriatic acid (20 mL 20 Baume diluted in 125 mL distilled water), then rinsed in distille d water and air dried. This procedure was very important because yeast cells can freely flow through these open pores and sometimes become stuck. This can result in clogging, which rest ricts the flow of anolyte through the electrode. Also th e mediator dyes used can stain the electrodes. Hardware Preparation: Biofuel cell structure was thoroughly washed using alcohol and distilled water. A new membrane, external tubing, valves, o-rings and syringes were used for each mediator.
34 Chapter 4 Experimental Results The results of the experimental program are shown in Figure 20. It should be noted that these results are with respect to a Ferricyanide cathode (Eo = +0.36 V), and that as a consequence the voltage and powe r-density data are lower than might be expected with an oxygen cathode (Eo = +0.82 V). As previously predicted, the results obta ined when NR was used alone were rather poor (in terms of peak power-density). It also proved to be very difficult to obtain consistent results, such that no performance curves were exactly alike, even after repeating the same NR experi ment eight times. Figure 20 shows a typical performance 0 0.2 0.4 0.6 0 0.1 0.20.30.4 0.5 0.6 Voltage [V] Power Density [W/m2] 1mM MB 0.5mM MB + 0.5mM NR 1mM NR Figure 20: Performance Results Using Single and Mixed Mediation.
35 curve for NR bounded by a shaded region that repr esents the spread of data obtained. It is believed that the data spread is indicative of the three possi ble pathways (n = 0, 1 or 2) elucidated in Figure 8, which might be utilized in countless combinations and permutations by the 1010 1011 individual yeast cells present in the MFC. No Pasteur Effect , the acceleration of metabolic rate during ferm entation, was observed with yeast under NR mediation which might have in creased the power by 3 to 4 times. Also since the concentration of sugar used in th ese experiments was low, no Crabtree effect was observed either. Clearly, MB was superior to NR when us ed as the sole mediator. The data was highly repeatable and showed power-density levels approximately four times that observed with NR. These re sults support the general hypot hesis that MB promotes anaerobic respiration, whereas NR is associated with fermentation only in yeast. When MB and NR were mixed, the affect on the MFC performance was dramatic, even though the overall mediator concentra tion remained unaltered. The peak powerdensity (0.52 W/m2) was 25% higher than with MB al one, and was very consistent and repeatable. This result was consistent with the hypothesi s presented in Figure 10, although more detailed stoichiometic meas urements will be needed for precise confirmation. In terms of peak power density the th ree cases of 1 mM NR, 1 mM MB, and mM NR mixed with mM MB were in th e ratio 15:69:84 respectively, which was very close to the ratio of predicted energy efficiencies of 15:65:87.
36 Chapter 5 Comparison of Different Electrodes Clearly, the performance of MFCs is intrinsically related to the electrode materials. The electrodes fo r the present work were cut fr om reticulated vitreous carbon (RVC) foam, of 100 pores per inch. This mate rial can be somewhat inconsistent in its manufacturing. Upon observation, the pores in the middle appeared darker and seemed to be denser. Electrodes #1 Electrodes #1 Electrodes #2 Electrodes #2 Figure 21: Location Where Electrodes We re Cut from Piece of RVC Foam Plate.
37 To study the variabil ity of RVC electrode s, two sets (anode and cathode) were cut from the same piece of material. The electrodes for the first set (labeled as electrodes #1) were cut close to the middle of the sheet, whereas, the second set (labeled as electrodes #2) were cut closer to the corners of the sheet. It might be expected that all electrodes, cut the same way from the same piece of ma terial, would perform the same, but this proved not to be the case. Figure 27, in Appendix B, shows MFC perf ormance using the two sets of RVC electrodes. Clearly, the Elec trodes #1 showed substantially superior performance in comparison to the Electrodes #2. In order to discover why the two sets of electrodes performed so differently, their conductivity was measured with an ohmmeter. It was found that the resistance was 2 k /inch for the Electrodes #2 and 200 /inch for Electrodes #1. Thus, a factor of 10 times in resistance was observed from the same piece of RVC material. This substantial difference could confirm why the peak power was so different between the two sets of electrodes. Based on the results of these electrode tests, it was Electrodes #1 that were employed for the mediator test results reported in Figure 20.
38 Chapter 6 Yeast Staining Phenomena Mediator dyes are able to penetrate the outer cell lipid membranes and plasma wall of the yeast even though th e yeast is so tiny, taking up a volume of less than 50 m3. Inside, the mediator interacts with the me tabolic pathway, accepting electrons from coenzyme intermediates, or acting as a surroga te terminal electron acc eptor in the absence of oxygen . In the latter case the process is more of an anaerobic respir ation, rather than a traditional fermentation. Upon leavi ng the living cell the mediator becomes reoxidized at the anode, ther eby providing a circulatory electron transport coupling mechanism . Yeast cells have disti nguishable features, or organelles, within the cell that behave like our organs. Organelles do different tasks, some br eak down food into required nutrients for use in their normal activities. Pa st research has shown that some organalles of yeast are affected by mediator dyes more than others. Tests us ing fluorochromic dyes show that particular dyes affect some in clusion bodies more than others, if not exclusively . With 1,000-fold magnificati on it is possible to ex amine yeast vacuole and cytosolic inclusion bodies to see how fluor chormic dyes affect them . Individual cells from a particular yeast strain of a single species can display morphological and color heterogeneity . On the macromolecular level, the body of yeast comprises proteins, glycoproteins, polysaccharides, polyphosphates, lipids, and nucleic acids . These constituents attract
39 these dyes and the dyes act in place of an enzyme when breaking molecules down in order to obtain ATP energy. From Table 2, it can be seen that Neutral Red Dye interacts with the Vacuoles only whereas Methylene Blue interacts within th e whole cell. This specificity of NR to the vacuoles of yeast may partly explain the rather poor performance observed with this mediator in the MFC. NR is known to perform mu ch better with e-coli probably because NR acts more generally in this bacterium. Dye Structures visualized Comments Methylene Blue Whole cell N on-viable cells stain blue Aminoacidine Cell walls Indi cator of surface potential F-C ConA Cell walls Binds specifically to mannan Calcofluor white Bud scars Chitin in scar fluoresces DAPI(4,6-diamidino-2-phenylindole)Nuclei DNA fluoresces Neutral Red Vacuoles Vacuoles stain red-purple Iodine Glycogen depositsGlycogen stained red-brown DAPI Mitochondria Mitochond ria fluoresce pink-white Rhodamine Mitochondria Unlike NR, MB is able to pass through the membrane of yeast and all other inclusion bodies of the whole cell. This suggests that MB works with all parts of the yeast cell acting as an enzyme substitute in many different locations of the TCA cycle, see Figure 6. Table 2: Structure-s p ecific D y es fo r Yeast Cells [ 20 ] .
40 Chapter 7 Hydrogen Peroxide A study on comparing different cathode oxidants was also performed, see Figure 29 in Appendix B. Some past researcher s have chosen Hydrogen Peroxide as the catholyte based on its ability to accept io ns and electrons. Hydrogen Peroxide (H2O2) combined with Hydrogen ions and electrons become water (H2O), which has a very low energy state. Using Hydrogen Peroxide in a MFC the cathodic reaction would become: H2O2 + 2 H+ + 2 e2 H2O The cathode redox potential in this case is +0.82 V compared to +0.36 V for Ferricyanide, which would inevitably lead to a larger OCV for the MFC. A test using a 1mM concentration of Met hylene Blue in the a nolyte was used to compare Hydrogen Peroxide to Ferricyanide. As expected the open circuit potential was greater when using the Hydrogen Peroxide, see Figure 34 in Appendix C. Hydrogen Peroxides main drawback was that its pow er dropped off much quicker than during the Ferricyanide tests. This is because the above cathodic reaction requires a catalyst to proceed at acceptable speed. A noble metal such as platinum is requi red as the catalyst, which is expensive and still requires elevated temperatures. So to maximize power density, ferricyanide instead of Hydrogen Pero xide was used for the entire test program reported here.
41 Ferricyanide also accepts electrons easily but its chemical kinetics are very fast. The cathodic reaction with ferricyanide is: Fe(CN)6 3+ H+ + eFe(CN)6 -4 + H+ The drawback of ferricyanide is that the cathode reacti on does not sweep up hydrogen ions (H+) such that the buffer soon becomes e xhausted with tan associated rise in pH.
42 Chapter 8 Conclusion Mediatorless biofuel cells are certainly desirable for a number of reasons (lower cost, flow-through design), but th e output power possible from such systems is inevitably rather low. Although certain ion-reducing bacter ia can transfer electr ons directly to an anode, they do this via an outer membra ne cytochrome-c of relatively high redox potential (+0.2 V). Because cytochrome-c (an electron carrier) is lo cated towards the end of the transport chain the ion-reducing bacter ia utilize approximately half of the fuel energy in ATP production. This results in an overall energy-efficiency significantly less than the 87% possible with ta rgeted mediators, as shown in the formulation on page 18. Despite the rather narrow approach a dopted for predicting performance, the results obtained from the experimental yeas t-catalyzed MFC mirrore d these expectations quite closely in terms of relative power-den sity. In terms of peak power density, the three cases of 1 mM NR, 1 mM MB, and mM NR mixed with mM MB were in the ratio 15:69:84 respectively, shown in Figure 22. This experimental data was very close to the ratio of predicted ener gy efficiencies of 15:65:87 cal culated on page 15, 17, and 18 respectively. The superior performance reported  for e.coli with NR alone was neither predicted nor observed with yeast. This is most likely due to the significant differences in fermentative pathways adopted by the two organisms. In addition, the mitochondrial physiology of yeast may impe de trans-cellular passage of certain
43 mediators. For example, it is well known that MB is able to stain yeast whole-cells, while NR exclusively stains the vacuoles. It is plausible that such dye specificity to certain physiological structures may impact th eir effectiveness as mediators. For powerdensity, a combination of mediators was best suited. The high energy efficiency and improved power density of mixed-mediated MF Cs makes them an attractive alternative to direct-exchange designs. However, mediated systems are currently best suited to biobattery applications where mediators are not lo st via the waste stream of flow-through biofuel cell configurations. Immobilization or mediator separation/recycling techniques may eventually remove this limitation. Future research is required to study me diator ratios other than 50% NR + 50% MB, such as 20% NR + 80% MB etc., in orde r to find the optimum mi x ratio. In addition the experiments need to be repeated with an oxygen cathode using Pl atinum as a cathodic catalyst.
44 44 Figure 22: Comparison of Peak Power. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 0.00000.10000.20000.30000.40000.50000.60000.7000Voltage [V]Power [mW] Neutral Red 50% NR + 50% MB Methylene Blue
45 References  Lewis, Nancy D. http://itp.nyu.edu/~ nql3186/electricity/timeline.html#Volta Accessed: November 2005.  Wilkinson, S., Gastrobots Benefits and Challenges of Microbial Fuel Cells in Food Powered Robot Applications, J. Autono mous Robots, 9(2), September, 2000, pp. 99111.  Wilkinson, S., Juice from Juice Electricity Generation for Carpophagous Gastrobots, IASTED 2000 Int. Conf. On Robotics & Applications, Clearwater, Florida, Nov. 19-22, 2001, paper #246-036.  Wilkinson, S., Gastrono me A Pioneering Food Powe red Mobile Robot, Proc. IASTED 2000 (8th) Int. Conf. On Robotics & App lications, Honolulu, Hawaii, Aug. 14-16, 2000, pp. 176-181.  Bennetto, H.P., Electricity Generation by Microorganisms", Biotechnology Education Vol. 1, No. 4, Column 3, 1990, p. 164.  Bennetto, H.P., Dew, M.E., Stirling, J.L ., and Tanaka, K., Rates of Reduction of Phenothiazine Redox Dyes by E. col i, Chemistry and Industr y, 1981, pp. 776778.  Delaney, G.M., Bennetto, H.P., Mason, J.R., Roller, H.D., Stirling, J.L., and Thurston, C.F., Electron-tran sfer Coupling in Microbial Fuel Cells: 2. Performance of Fuel Cells Containing Selected Micoor ganism-Mediator-Substrate Combinations, J. Chem.Tech. Biotechnol ., 34B, 1984, pp.1327.  Bennetto, H.P., Microbial Fuel Cells, Life Chemistry Reports, Harwood Academic Publishers GmbH and OPA Ltd, Vol. 2, No. 4., 1984, pp. 371-373.  Lithgow, A.M., Romero, L., Sanchez, I.C ., Souto, F.A., and Vega, C.A., Interception of Electron-transport Chain in Bacter ia with Hydrophilic Redox Mediators, J. Chem. Res. (S), 1986, pp. 178179.  Tanaka, K., Vega, C.A., and Tamamu shi, R., Thionine and Ferric Chelate Compounds as Coupled Mediators in Microbial Fuel Cells, Bioelectricity and Bioenergetic s, 11, 1983, pp. 289297.
46  Bennetto, H.P., Stirling, J. L., Tanaka, K., Vega, C.A., Soc. Gen. Microb. Quart., Vol. 8, No. 1, 1980, p. 37; Anodic Reacti ons in Microbial Fuel Cells Biotechnol. Bioeng. Vol. 25, 1983, pp. 559-568.  Zhang, X. and Halme, A., A Summary of the Study of Bio-electrochemical Fuel Cell by Using Saccharomyces cerevisia e, Research Reports of Automation Technology Laboratory of HUT, No. 10, January, 1994.  Park, D.H. and Zeikus, J.G., Electric ity Generation in Microbial Fuel Cells Using Neutral Red as an Electronophore, Appl. Environ. Microbiol. 66, 2000, pp. 12921297.  Clark, W.M., Oxidation-Reduction Potentials of Organic Systems, Williams & Wilkins, Baltimore, 1960, pp. 125, 131, 132, 415 and 496.  Park, D.H. and Zeikus, J.G., Utilization of Electrically Reduced Neutral Red by Actinobacillus succinogenes: Physiologica l Function of Neutral Red in MembraneDriven Fumarate Reduction and Energy Conservation, J. Bacteriology Vol. 181, No. 8, April 1999, pp. 2403-2410.  Thunberg, T., Das Reduktions-Oxydati onspotential eines Gemisches von SuccinatFumerat, Skand. Arch. Physiol ., Vol. 46, 1925, pp. 339.  Quastel, J.H. and Whetham, M.D., The Equilibria Existing Between Succinic, Fumaric and Malic Acids in the Pr esence of Resting Bacteria, B iochem. J ., Vol. 18, 1924, p. 519.  Thurston, C.F., Bennetto, H.P., Delaney, G.M., Mason, J.R., Roller, H.D., and Stirling, J.L., Glucose Metabolism in a Mi crobial Fuel Cell Stoichiometry of Product Formation in a Thionine-mediated Proteus vulgaris Fuel Cell and its Relation to Coulombic Yields, Journal of General Microbiolog y, 131, 1985, pp. 13931401.  Gottschalk, G., Bacterial Metabolism, 2nd Edition, Springer-Verlag, Ch. 8., Section IB, 1986, p. 213.  Feldman, H., Yeast Cell Architectur e and Function, Butendadt Institute, University of Munich, 2005, h ttp://biochemie.web.med.unimuenchen.de/Yeast_Biol/ Accessed: December 2005.
47 Bibliography Bungay, Rensselaer Polytechnical Institute. http://www.rpi.edu/dept/chem-eng/Bi otech-Environ/Con tin/overload.html Dickinson, J.R. and Schweizer, M., eds., The Metabolism and Molecular Physiology of Saccharomyces cerevisiae , Taylor & Francis, London/Philadelphia, 1999. Moller, P., Electric Fishes: History and Behavior, Fish a nd Fisheries Series 17, Chapman & Hall.NY, 1995. Wilkinson, S. and Guanio, D., Development of a Spectrometric Citrus Odor Sensor for a Frugivorous Robot, Proc. 6th IASTED Int. Conf. Robotics & Manufacturing, paper 284-014, Banff, Canada, July 26-29, 1998, pp. 252-255.
49 Appendix A: Tabulated Experimental Results Table 3: Experimental Results for NR Only Electrodes #2. Resistance Volts Power Curr ent LOG(I) ResistanceVolts Power Current LOG(I) V mW Amp Amp V mW Amp Amp 1000080 0.3603 0.0001 3.60E-07-6.4434 1080 0.32860.1000 3.04E-04 -3.5168 900080 0.3603 0.0001 4.00E-07-6.3976 980 0.32270.1063 3.29E-04 -3.4824 800080 0.3604 0.0002 4.50E-07-6.3463 880 0.31620.1136 3.59E-04 -3.4445 700080 0.3603 0.0002 5.15E-07-6.2885 780 0.31030.1234 3.98E-04 -3.4003 600080 0.3604 0.0002 6.01E-07-6.2214 680 0.30340.1354 4.46E-04 -3.3505 500080 0.3605 0.0003 7.21E-07-6.1421 580 0.29400.1490 3.04E-04 -6.4434 400080 0.3605 0.0003 9.01E-07-6.0452 480 0.28400.1680 3.04E-04 -6.4434 300080 0.3606 0.0004 1.20E-06-5.9202 380 0.26600.1862 3.04E-04 -6.4434 200080 0.3605 0.0006 1.80E-06-5.7443 280 0.24300.2109 3.04E-04 -6.4434 100080 0.3603 0.0013 3.60E-06-5.4437 180 0.20500.2335 3.04E-04 -6.4434 90080 0.3603 0.0014 4.00E-06-5.3980 170 0.19100.2146 3.04E-04 -6.4434 80080 0.3602 0.0016 4.50E-06-5.3470 160 0.18500.2139 3.04E-04 -6.4434 70080 0.3602 0.0019 5.14E-06-5.2891 150 0.17400.2018 3.04E-04 -6.4434 60080 0.3601 0.0022 5.99E-06-5.2223 140 0.16500.1945 3.04E-04 -6.4434 50080 0.3601 0.0026 7.19E-06-5.1432 130 0.15500.1848 3.04E-04 -6.4434 40080 0.3599 0.0032 8.98E-06-5.0467 120 0.14600.1776 3.04E-04 -6.4434 30080 0.3596 0.0043 1.20E-05-4.9225 110 0.13700.1706 3.04E-04 -6.4434 20080 0.3591 0.0064 1.79E-05-4.7475 100 0.12700.1613 3.04E-04 -6.4434 10080 0.3575 0.0127 3.55E-05-4.4502 90 0.11600.1495 3.04E-04 -6.4434 9080 0.3567 0.0140 3.93E-05-4.4058 80 0.10600.1405 3.04E-04 -6.4434 8080 0.3560 0.0157 4.41E-05-4.3560 70 0.09500.1289 3.04E-04 -6.4434 7080 0.3552 0.0178 5.02E-05-4.2996 60 0.08500.1204 3.04E-04 -6.4434 6080 0.3542 0.0206 5.83E-05-4.2347 50 0.07200.1037 3.04E-04 -6.4434 5080 0.3528 0.0245 6.94E-05-4.1583 40 0.06000.0900 3.04E-04 -6.4434 4080 0.3508 0.0302 8.60E-05-4.0656 30 0.04800.0768 3.04E-04 -6.4434 3080 0.3482 0.0394 1.13E-04-3.9467 20 0.03200.0512 3.04E-04 -6.4434 2080 0.3424 0.0564 1.65E-04-3.7835 10 0.01700.0289 3.04E-04 -6.4434
50 Appendix A: (continued) Table 4: Experimental Results for NR and MB Electrodes # 2. Resistance Volts Power Cu rrent LOG(I)ResistanceVo lts Power Current LOG(I) [ ] [V] [mW] [Amp] [ ] [V] [mW] [Amp] 1,000,080 0.5922 0.0004 5.92E-07-6.22761,080 0.54490.2749 5.05E-04 -3.2971 900,080 0.5921 0.0004 6.58E-07-6.1819980 0.53850.2959 5.49E-04 -3.2600 800,080 0.5921 0.0004 7.40E-07-6.1307880 0.53210.3217 6.05E-04 -3.2185 700,080 0.5921 0.0005 8.46E-07-6.0728780 0.52450.3527 6.72E-04 -3.1723 600,080 0.5921 0.0006 9.87E-07-6.0058680 0.51500.3900 7.57E-04 -3.1207 500,080 0.5921 0.0007 1.18E-06-5.9266580 0.50350.4371 8.68E-04 -3.0614 400,080 0.5921 0.0009 1.48E-06-5.8298480 0.48910.4984 1.02E-03 -2.9918 300,080 0.5920 0.0012 1.97E-06-5.7049380 0.47030.5821 1.24E-03 -2.9074 200,080 0.5919 0.0018 2.96E-06-5.5290280 0.43410.6730 1.55E-03 -2.8096 100,080 0.5916 0.0035 5.91E-06-5.2283180 0.37640.7871 2.09E-03 -2.6796 90,080 0.5915 0.0039 6.57E-06-5.1827170 0.36490.7832 2.15E-03 -2.6683 80,080 0.5914 0.0044 7.39E-06-5.1316160 0.35380.7823 2.21E-03 -2.6554 70,080 0.5913 0.0050 8.44E-06-5.0738150 0.34310.7848 2.29E-03 -2.6407 60,080 0.5912 0.0058 9.84E-06-5.0070140 0.33160.7854 2.37E-03 -2.6255 50,080 0.5910 0.0070 1.18E-05-4.9281130 0.31900.7828 2.45E-03 -2.6102 40,080 0.5907 0.0087 1.47E-05-4.8316120 0.30650.7829 2.55E-03 -2.5928 30,080 0.5902 0.0116 1.96E-05-4.7073110 0.29500.7911 2.68E-03 -2.5716 20,080 0.5900 0.0173 2.94E-05-4.5319100 0.27700.7673 2.77E-03 -2.5575 10,080 0.5867 0.0341 5.82E-05-4.235090 0.26100.7569 2.90E-03 -2.5376 9,080 0.5859 0.0378 6.45E-05-4.190380 0.24100.7260 3.01E-03 -2.5211 8,080 0.5850 0.0424 7.24E-05-4.140370 0.22200.7041 3.17E-03 -2.4987 7,080 0.5841 0.0482 8.25E-05-4.083560 0.20000.6667 3.33E-03 -2.4771 6,080 0.5825 0.0558 9.58E-05-4.018650 0.17500.6125 3.50E-03 -2.4559 5,080 0.5807 0.0664 1.14E-04-3.941940 0.14600.5329 3.65E-03 -2.4377 4,080 0.5780 0.0819 1.42E-04-3.848730 0.11500.4408 3.83E-03 -2.4164 3,080 0.5740 0.1070 1.86E-04-3.729620 0.07900.3121 3.95E-03 -2.4034 2,080 0.5659 0.1540 2.72E-04-3.565310 0.04100.1681 4.10E-03 -2.3872
51 Appendix A: (continued) Table 5: Experimental Results for MB Only Electrodes #2. Resistance Volts Power Cu rrent LOG(I)ResistanceVo lts Power Current LOG(I) V mW Amp Amp V mW Amp Amp 1000080 0.5524 0.0003 5.52E-07-6.25781080 0.5306 0.2607 4.91E-04 -3.3087 900080 0.5525 0.0003 6.14E-07-6.2119980 0.5277 0.2842 5.38E-04 -3.2688 800080 0.5525 0.0004 6.91E-07-6.1608880 0.5248 0.3130 5.96E-04 -3.2245 700080 0.5525 0.0004 7.89E-07-6.1028780 0.5212 0.3483 6.68E-04 -3.1751 600080 0.5526 0.0005 9.21E-07-6.0358680 0.5167 0.3926 7.60E-04 -3.1193 500080 0.5526 0.0006 1.11E-06-5.9566580 0.5107 0.4497 8.81E-04 -3.0553 400080 0.5527 0.0008 1.38E-06-5.8597480 0.5026 0.5263 1.05E-03 -2.9800 300080 0.5527 0.0010 1.84E-06-5.7347380 0.4901 0.6321 1.29E-03 -2.8895 200080 0.5527 0.0015 2.76E-06-5.5587280 0.4718 0.7950 1.69E-03 -2.7734 100080 0.5526 0.0031 5.52E-06-5.2579180 0.4350 1.0513 2.42E-03 -2.6168 90080 0.5526 0.0034 6.13E-06-5.2122170 0.4275 1.0750 2.51E-03 -2.5995 80080 0.5526 0.0038 6.90E-06-5.1611160 0.4209 1.1072 2.63E-03 -2.5799 70080 0.5526 0.0044 7.89E-06-5.1032150 0.4124 1.1338 2.75E-03 -2.5608 60080 0.5526 0.0051 9.20E-06-5.0363140 0.4027 1.1583 2.88E-03 -2.5411 50080 0.5526 0.0061 1.10E-05-4.9573130 0.3932 1.1893 3.02E-03 -2.5193 40080 0.5525 0.0076 1.38E-05-4.8606120 0.3812 1.2109 3.18E-03 -2.4980 30080 0.5523 0.0101 1.84E-05-4.7361110 0.3676 1.2285 3.34E-03 -2.4760 20080 0.5519 0.0152 2.75E-05-4.5609100 0.3510 1.2320 3.51E-03 -2.4547 10080 0.5507 0.0301 5.46E-05-4.262590 0.3324 1.2277 3.69E-03 -2.4326 9080 0.5504 0.0334 6.06E-05-4.217480 0.3136 1.2293 3.92E-03 -2.4067 8080 0.5501 0.0375 6.81E-05-4.167070 0.2900 1.2014 4.14E-03 -2.3827 7080 0.5497 0.0427 7.76E-05-4.109960 0.2620 1.1441 4.37E-03 -2.3598 6080 0.5491 0.0496 9.03E-05-4.044350 0.2320 1.0765 4.64E-03 -2.3335 5080 0.5483 0.0592 1.08E-04-3.966840 0.2010 1.0100 5.03E-03 -2.2989 4080 0.5471 0.0734 1.34E-04-3.872630 0.1580 0.8321 5.27E-03 -2.2785 3080 0.5451 0.0965 1.77E-04-3.752120 0.1110 0.6161 5.55E-03 -2.2557 2080 0.5412 0.1408 2.60E-04-3.584710 0.0610 0.3721 6.10E-03 -2.2147
52 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 05001000150020002500Load Resistance [ ]Terminal Voltage [V] Neutral Red 50% NR +50% MB Methylene Blue 52 Figure 23: Voltage vs. Resistance. Appendix B: Graphical Experimental Results
53 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 05001000150020002500 Load Resistance [ohm]Power Output [mW] 50% NR + 50% MB Methylene Blue Neutral Red 53 Figure 24: Power vs. Resistance. A pp endix B: ( Continued )
54 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 0.00.10.20.126.96.36.199.7 Voltage [V]Power [mW] Neutral Red 50% NR + 50% MB Methylene Blue 54 Figure 25: Power vs. Voltage. A pp endix B: ( Continued )
55 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 -6-5-4-3-2-10 Logarithm of Current [log A]Terminal Voltage [V] 5 MB + 5 NR Neutral Red Methylene Blue 55 Figure 26: Log of Current vs. Voltage. A pp endix B: ( Continued )
56 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 1.2000 1.4000 1.6000 00.10.20.30.188.8.131.52 Voltage [V]Power [mW] Neutral Red Neutral Red New Electrodes 50% NR + 50% MB 50% NR + 50% MB New Electrodes Methylene Blue Methylene Blue New Electrodes Figure 27: Power vs. Voltage for Different Electrodes. 56 A pp endix B: ( Continued )
57 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 05001000150020002500 Load Resistance [ohm]Power Output [mW] Neutral Red Electrodes #1 50% NR + 50% MB Electrodes #2 50% NR + 50% MB Electrodes #1 Methylene Blue Electrodes #2 Methylene Blue Electrodes #1 Neutral Red Electrodes #1 57 Figure 28: Power vs. Resistan ce for Different Electrodes. A pp endix B: ( Continued )
58 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 00.10.20.30.184.108.40.206Voltage [V]Power [mW] MB with Ferricyanide in the Cathode MB with H2O2 in the Cathode 58 Figure 29: Comparison of Hydrogen Peroxide to Ferricyanide in the Cathode. A pp endix B: ( Continued )
59 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0:000:010:020:030:040:050:06Time [hr:min]Voltage [V] 59 Figure 30: Fluke Data for No Mediator. A pp endix C: Fluke Data
60 60 Figure 31: Fluke Data for MB. 0.0000 0.1000 0.2000 0.3000 0.4000 0.5000 0.600000:00 01:15 02:30 03:45 05:00 06:15 07:30 08:45 10:00 11:15 12:30 13:45 15:00 16:15 17:30 18:45 20:00 21:15 22:30 23:45 25:00 26:15 27:30 28:45 30:00 31:15 32:30 33:45 35:00 36:15 37:30 38:45 40:00 41:15 42:30 43:45 45:00Time [hr:min:sec]Voltage [ V A pp endix C: ( Continued )
61 0.0000 0.1000 0.2000 0.3000 0.4000 0.5000 0.6000 0.7000 0:000:050:100:150:200:250:300:350:400:45Time [hr:min]Voltage [V] 61 Figure 32: Fluke Data for MB and NR. A pp endix C: ( Continued )
62 0.0000 0.0500 0.1000 0.1500 0.2000 0.2500 0.3000 0.3500 0.40000 :00 0:00 0 :0 1 0 :0 2 0 :03 0:04 0 :0 5 0 :0 6 0 :07 0 :0 8 0 :0 9 0 :1 0 0 :11 0: 1 1 0 :1 2 0 :1 3 0:14 0: 1 5 0 :1 6 0 :1 7 0:18 0:19 0 :2 0 0 :21 0:22 0:22 0 :2 3 0 :24 0:25 0:26 0 :27 0 :28 0:29Time [hr:min]Voltage [V] 62 Figure 33: Fluke Data for NR. A pp endix C: ( Continued )
63 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0:000:010:020:030:04Time [hr:min]Voltage [V] 63 Figure 34: Fluke Data for MB With H2O2 in the Cathode. A pp endix C: ( Continued )
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001795193
007 cr mnu|||uuuuu
008 070629s2006 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001544
Klar, Jason C.
Optimizing biofuel cell performance using a targeted mixed mediator combination
h [electronic resource] /
by Jason C. Klar.
[Tampa, Fla] :
b University of South Florida,
ABSTRACT: A study of how mediators interact with the catabolic pathways of microbes was undertaken with a view towards improving the performance of microbial fuel cells. The use of mediators is known to improve the power density in microbial fuel cells, but this work suggests that no single mediator is ideally suited to the task. Instead, a carefully selected mixture of two targeted mediators (Methylene Blue and Neutral Red) might be optimal. To test this hypothesis, a yeast-catalyzed microbial fuel cell was built and empirically evaluated under different mediation conditions while keeping all other parameters constant. The results clearly show that an appropriate mix of the two mediators mentioned could indeed achieve significantly superior performance, in terms of power-density, than when either mediator is used singly. All tests were carried out using the same overall mediator concentration.
Thesis (M.A.)--University of South Florida, 2006.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 63 pages.
Adviser: Stuart Wilkinson, Ph.D.
Microbial fuel cells.
x Mechanical Engineering
t USF Electronic Theses and Dissertations.