Citation
Effects of resistance, frequency, duty ratio on microbial fuel cells ( MFC ) performance

Material Information

Title:
Effects of resistance, frequency, duty ratio on microbial fuel cells ( MFC ) performance
Creator:
Feng, Shuo
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English

Thesis/Dissertation Information

Degree:
Master's ( Master of science)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
Department of Electrical Engineering, CU Denver
Degree Disciplines:
Electrical engineering
Committee Chair:
Park, Jaedo
Committee Members:
Roane, Timberley
Lee, Jung-Jae

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
Copyright Feng Shuo. Permission granted to University of Colorado Denver to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

Downloads

This item has the following downloads:


Full Text
EFFECTS OF RESISTANCE, FREQUENCY, DUTY RATIO ON MICROBIAL
FUEL CELLS (MFC) PERFORMANCE by
SHUO FENG
BS, Beijing University of Technology, 2014
A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Master of Science Electrical Engineering Program
2017


This thesis for the Master of Science degree by Shuo Feng
has been approved for the Department of Electrical Engineering Program
by
Jaedo Park, Chair Timberley Roane Jung-Jae Lee
July 12th, 2017
11


Feng, Shuo (M.S., Electrical Engineering Program)
Effects of Resistance, Frequency, Duty Ratio on Microbial Fuel Cells (MFC) Performance
Thesis directed by Associate Professor Jaedo Park
ABSTRACT
The performance of a single chamber batch feed microbial fuel cell (MFC) was investigated under different external load conditions. The combination of four different resistances (125 Q, 250 Q, 375 Q, 500 Q), four different duty ratios (0.25, 0.50, 0.75, 1.00) and two different switching frequencies (2000 Hz, 10000 Hz) were tested during the experiment. Four bio-chemical variables were also monitored during the experiment, including pH, electric conductivity (EC), dissolved oxygen (DO), Oxidation- Reduction potential (ORP). The relations between MFC voltage, power, and energy outputs and load conditions were discussed, lower power and energy production were found when the loads were not fixed. The relations between MFC four bio-chemical variables and load conditions were also discussed. The pH, DO, ORP values changes were not related to the load conditions changes, and longer energy extraction time and higher EC value were found when the switching frequency was higher. No other clear relations were found during the experiment. It shows the electrical stimulations may affect the bio-chemical activity inside the MFC reactor, but not in the way of pH, DO and ORP.
The form and content of this abstract are approved. I recommend its publication.
Approved: Jaedo Park


This file is dedicated to my family, who support me to finish writing my thesis and
support me for the Master program.
IV


ACKNOWLEDGMENTS
This thesis would not have been possible without the generous support of Dr. Park, Dr. Roane and Dr. Lee. Also, I would like to thank you Alaraj Muhannad, Babaiahgari Bhanu and all other members in ECRF group for the help throughout the whole proj ect.
v


TABLE OF CONTENTS
CHAPTER
I. INTRODUCTION...........................................................1
1.1 Microbial Fuel Cells for Electricity Generation..................1
1.2 Microbial Fuel Cells Configurations and Materials................2
1.2.1 Single-chamber reactors...................................2
1.2.2 Two-chambers reactors.....................................3
1.2.3 Upflow, Tubular type MFC (UMFC)...........................4
1.2.4 Sediments reactor.........................................5
1.2.5 MFC Stacks................................................6
1.2.6 Chamber materials.........................................6
1.2.7 Anode materials...........................................6
1.2.8 Cathode materials.........................................7
1.2.9 Membrane..................................................7
1.3 Current MFC Research Topics......................................7
1.3.1 Load and MFCs.............................................7
1.3.2 pH and MFCs...............................................8
vi


1.3.3 Electric Conductivity and MFCs
8
1.3.4 Dissolved Oxygen and MFCs.................................9
1.4 Objectives......................................................9
1.4.1 Objective I..............................................10
1.4.2 Objective II.............................................10
1.5 Organization of Thesis.........................................10
II. MFC CONFIGURATION AND OPERATIONS.....................................11
2.1 Single-chamber Design and Materials of Construction............11
2.2 Inoculum of MFC and Feed.......................................11
2.3 MFC Operation..................................................12
III. EXPERIMENT COMPONENTS AND METHODS...................................13
3.1 Test System....................................................13
3.2 MFC Reactor....................................................13
3.3 Probes.........................................................14
3.3.1 pH Probe.................................................15
3.3.2 C onductivity Prob e.....................................15
3.3.3 Dissolved Oxygen Probe...................................15
3.3.4 Oxidation-Reduction potential Probe......................16
vii


3.4 Control Board
16
3.5 Test and Analysis Method........................................18
IV. EXPERIMENT AM) RESULTS................................................20
4.1 MFC Connected with Same Equivalent Load.........................20
4.1.1 Voltage Output during MFC Operation.......................21
4.1.2 Power Output during MFC Operation.........................22
4.1.3 Energy Output during MFC Operation........................22
4.1.4 Solution pH Value during MFC Operation....................26
4.1.5 Solution Conductivity Value during MFC Operation..........28
4.1.6 DO Value during MFC Operation.............................31
4.1.7 ORP Value during MFC Operation............................33
4.1.8 Relation between Duty Switching Frequency and Four Bio- chemical
Values....................................................35
4.2 MFC Connected with Loads Applied Different Switching Frequency34
4.2.1 Voltage Output during MFC Operation..........................37
4.2.2 Power Output during MFC Operation............................39
4.2.3 Energy Output during MFC Operation...........................36
vm
4.2.4 Biochemical Values Output during MFC Operation
40


4.3 MFC under Different Operating Points.....................44
4.3.1 MFC Voltage and Power Output.......................45
4.3.2 MFC Biochemical Parameters.........................48
V. CONCLUSION......................................52
REFERENCE........................................................53
IX


TABLES
TABLE
4.1 Different Load Condition with Same Operating Point.....................20
4.2 Different Load Condition with Same Switching Frequency.................38
4.3 Different Load Condition under Different Operating Points..............44
4.4 Duty ratio, switchingfrequency and equivalent resistance values........46
4.5 Biochemical Values under Different Load Conditions.....................48
x


FIGURES
FIGURE
1.1 General Operating principles of a MFC...............................2
1.2 Schematic Diagram of Single-chamber MFC.............................3
1.3 Schematic Diagram of Two-chambers MFC...............................4
1.4 Schematic Diagram of Upflow MFC.....................................5
1.5 Schematic Diagram of Sediments MFC..................................5
2.1 MFC Reactor with Sensor Block......................................11
3.1 MFC Reactor Test System............................................13
3.2 Reactor during Inoculation.........................................14
3.3 Probes in Reactor..................................................14
3.4 Control Board......................................................16
3.5 Schematic of Converter.............................................17
3.6 Fixed Load.........................................................17
3.7 Load with Switching................................................18
3.8 Hard Switching Voltage Reading.....................................19
4.1 MFC Voltage under Same Equivalent Load.............................21
4.2 MFC Power under Same Equivalent Load...............................22
xi


4.3 MFC Energy under Same Equivalent Load.
23
4.4 MFC Energy under Same Equivalent 2000 Hz Load.........................24
4.5 MFC Energy under Same Equivalent 10000 Hz Load........................24
4.6 MFC pH under Same Equivalent Load.....................................26
4.7 MFC pH under Same Equivalent 2000 Hz Load.............................27
4.8 MFC pH under Same Equivalent 10000 Hz Load............................27
4.9 MFC EC under Same Equivalent Load.....................................28
4.10 MFC EC under Same Equivalent 2000 Hz Load............................29
4.11 MFC EC under Same Equivalent 10000 Hz Load............................30
4.12 MFC DO under Same Equivalent Load....................................31
4.13 MFC DO under Same Equivalent 2000 Hz Load............................32
4.14 MFC DO under Same Equivalent 10000 Hz Load............................32
4.15 MFC ORP under Same Equivalent Load...................................33
4.16 MFC ORP under Same Equivalent 2000 Hz Load............................33
4.17 MFC ORP under Same Equivalent 10000 Hz Load..........................34
4.18 Average MFC pH under Same Equivalent Load............................35
4.19 Average MFC EC under Same Equivalent Load............................36
4.20 Average MFC DO under Same Equivalent Load............................36
xii


4.21 Average MFC ORP under Same Equivalent Load
37
4.22 MFC Voltage under Same Load Duty Ratio..............................39
4.23 MFC Power under Same Load Duty Ratio................................40
4.24 MFC Energy under Same Load Duty Ratio...............................40
4.25 MFC Energy Chart under Same Load Duty Ratio.........................41
4.26 MFC pH under Same Load Duty Ratio...................................42
4.27 MFC EC under Same Load Duty Ratio...................................42
4.28 MFC DO under Same Load Duty Ratio...................................43
4.29 MFC ORP under Same Load Duty Ratio..................................43
4.30 MFC Polarization Curve..............................................45
4.31 MFC Polarization Curve Part.........................................46
4.32 MFC Test Data Curve................................................47
4.33 Average MFC pH under Different Equivalent Load......................49
4.34 Average MFC EC under Different Equivalent Load......................49
4.35 Average MFC DO under Different Equivalent Load......................50
xm
4.36 Average MFC ORP under Different Equivalent Load.
50


CHAPTER I
INTRODUCTION
1.1 Microbial Fuel Cells for Electricity Generation
Microbial fuel cell (MFC) is a rapidly growing research topic in recent decades for its electricity generation capability. An MFC reactor is a bio-electro-chemical system and an Oxidation-reduction reaction is happening in the reactor that use bacteria as the bio-catalysts [1][2], It is a complex system involving the knowledge from different scientific and engineering fields and to analyze the performance of an MFC system as an energy source, the knowledge from microbiology, electro-chemistry and electrical engineering are required. In the MFCs systems, bacteria produce electrons by consuming the substrates existing in solution and transfer them to the anode (negative terminal) and electrons can flow to the cathode (positive terminal) through a load, such as a resistor that is connected outside the reactor [2], The current that generated by the MFC reactor is flowing from cathode to anode which is opposite to the electron flow. There are three ways that electrons can be transferred to the anode discovered. First is by electron mediators or shuttles [3], second is by direct membrane associated electron transfer [4], and the third way is by so-called nanowires [5] produced by the bacteria, even though the mechanism of electron transfer is not well known [6], There are various microbial cultures that are known to generate electricity in the MFCs. The research shows that in general MFCs using mixed cultures usually achieve substantially higher power densities than those with pure cultures [7], Theoretically any biodegradable organic matter can be used to build an MFC reactor, including acids, alcohols, carbohydrates, proteins, and cellulose [8][9], Various oxidants for the reduction reaction at cathode (electron acceptor) have been used in MFCs, such as oxygen, nitrate, ferricyanide, permanganate, ferric iron [10], Wastewater and oxygen are the most promising
1


electron donor and acceptor, respectively, as they are the most accessible and economical [9], In typical air-cathode MFC reactors the electrons will combine at the cathode with the protons that come from the anode and oxygen provided from air; so the side product is just water.
1.2 Microbial Fuel Cells Configurations and Materials MFCs are being constructed using a variety of materials, and in an increasing diversity of configurations. The most common designs include the single-chamber air cathode reactor, two-chamber reactor used in the research laboratory and sediments reactor used in the field. The general operating principle of a MFC reactor is showed in Figure 1.1 [10], and some of the MFC designs are discussed in this section.
Figure 1.1: General Operating principles of a MFC.
1.2.1 Single-chamber Reactors
Consisting only the anode chamber, in this configuration, the anode and cathode are placed on either side of the chamber, usually cube shape. The cathode is exposed to air on one side and the solution on the other side, inside of the anode chamber. The proton exchange membrane (PEM) used in this system primarily to keep water from leaking through the cathode, and it also reduces
2


oxygen diffusion into the anode chamber. The anode side, usually a brush where the bacteria
♦ Electrons *
Oxidabon/reduction between enzyme and substrate
*
Carbon paper (anode)
Inoculant
4
M- •*'
Substrate
2v ♦ 2H* ♦ 0 => H,0
Inside chamber
Carbon paper w/ platinum exposed to air (cathode)
Figure 1.2: Schematic Diagram of Single-chamber MFC. grow, is also sealed in the solution, so that air cannot enter. The single-chamber, air-cathode
designs reactors have higher power density output relative to two-chamber reactors, due to its low internal resistance [11], One of the great challenges for the PEM-less MFC is that the Coulombic efficiency is much lower than those with PEM [12], A schematic diagram of singlechamber MFC reactor [13] is showed in Figure 1.2.
1.2.2 Two-chambers Reactors
Consisting two containers, usually two bottles connected by tube, or two cubes connected side by side, containing a separator in the middle, which is usually a PEM such as Nafion or Ultrex, or a plain salt bridge [14] [15] [16], The important point to this design is using a membrane that allows only protons, not allowing the substrate or electron to pass between the chambers. The amount of power that generating from these systems is affected by the surface area of the cathode relative to that of the anode [17], The membrane and longer distance between anode and cathode cause a higher internal resistance within a two-chamber MFC reactor [18], which result in lower
3


power density output compare to single-chamber reactors. A schematic diagram of two-chamber MFC reactor [19] is showed in Figure 1.3
Figure 1.3: Schematic Diagram of Two-chambers MFC.
1.2.3 Upflow, Tubular type MFC (UMFC)
There are mainly two kinds of upflow, tubular type MFC reactors. One is just like a two-chambers reactor, the anode chamber is put under the cathode chamber, separated by a PEM membrane. The other kind is comprised of a PEM membrane, which was folded and sealed through soldering to provide a cylindrical structure. Two glass stoppers, each with two inlets/outlets were inserted on both sides of this cylinder [20], The advantage of this design is that it can reduce the anode cathode electrode spacing. A schematic diagram of upflow MFC reactor [20] is showed in Figure 1.4.
4


Reference
electrode
Air pump
Resistor
Cathode Chamber
^L,
I Effluent
i i
Reference electrode C
\
Separator
Anode Chamber
3—©—^
Influent
Figure 1.4: Schematic Diagram ofUpflowMFC.
1.2.4 Sediments Reactor
It is one of the MFC applications, using carbon cloth as anode, which was buried below and parallel to the surface of the sediment. The cathode was positioned in the seawater, above and parallel to the sediment surface. Typically, the seawater layer above the cathode was 5 cm in depth [21], No separator was used in the cell. So this is a membrane less system. The anode and cathode were maintained as horizontal positions. A schematic diagram of Sediments MFC reactor [21] is showed in Figure 1.5.
Figure 1.5: Schematic Diagram of Sediments MFC.
5


1.2.5 MFC Stacks
Any of those reactor configurations can be used to build an MFC stack. In most cases, the MFC stack was constructed on the basis of tubular air-cathode MFCs. Usually individual cells were electrically connected in series by wires and hydraulically joined by conductive substrate flow, the performance degradation phenomenon was observed [22], The open circuit voltage (OCV) and current flow of stacked MFC were also lower [22],
1.2.6 Chamber Materials
The materials used to build up the reactor chamber must be non-toxic to the bacteria. Polyvinylchloride (PVC) plastic tube and Plexiglas are two common materials since they are easy to find, easy to make to any shape and at a low cost.
1.2.7 Anode Materials
Anodic materials are required to be conductive, biocompatible, and chemically stable in the reactor solution, but copper is not useful due to the copper ions are toxic to bacteria [10], Since carbon materials are well suited for bacterial growth and have high conductivity, so most of the anode materials are carbon-based, including carbon paper, carbon cloth, reticulated vitreous carbon (RVC), graphite fiber brush [23], Carbon paper and carbon cloth are the most flexible materials among those but very expensive. The RVC has been ideal for the continuous up flow MFCs because it has excellent conductivity. The increasing accessible surface area increased current density, so the graphite fiber brush was developed for the MFCs [23], Graphite brush electrodes can have an extremely high surface area by just having a small diameter [24], During the experiments, we also found out the plastic brush with the iron rod after an acid reaction showed good performance for the bacteria.
6


1.2.8 Cathode Materials
The same materials that have been described for anodes have also been used as cathodes, and different oxidants have been used as the electron acceptor in MFCs. Ferricyanide (Ks/I'efCN)6j) is very popular as an experimental electron acceptor in microbial fuel cells [25], and oxygen is the most suitable electron acceptor for the MFC since its high and oxidation potential and it is everywhere and the free of cost [10],
1.2.9 Membrane
The membrane in MFC reactor designs is often used as the separation of the anode and the cathode compartments, for example a PEM. The most commonly used membranes are made by Nafion. The use of membranes in MFCs increases internal resistance that decreases power production [18], Alternatives to Nafion, such as Ultrex CMI also are well suited for MFC reactor designs at a lower cost [10],
1.3 Current MFC Research Topics
MFC performance can be affected by many different aspects, including the factors like circuit load, internal resistance, solution pH value, temperature and others. In the view point of the electrical engineering, the performance of a MFC reactor is evaluated by its power generating capability, so the literature researches were mainly focused on MFC power output different conditions.
1.3.1 Load and MFCs
Differences in the external load (resistance) were shown to lead to some significant differences in MFC cell operation. The relation between different external load and MFC reactor current generation and power performance were investigated. Most of those tests were conducted after connected different reactors with same configuration to different load for a long period of time,
7


recorded data hourly. It is considered that with different fixed resistance load, the bacterial community around the anode are formed differently, higher external load lead to higher biomass growth, lower current generating [26], Also the increased external load can cause less electroactive bacteria growth, which result in lower voltage output [27], Although different bacteria community were formed, no significant power output differences were found, according to the power curve. The maximum power density points were similar, even the inoculum preparations were different [28],
1.3.2 pH and MFCs
The internal pH environment of an MFC reactor is one of the important factor for MFC power output, which can influence substrate metabolic activity and impact e and H+ generation process, due to that most bacteria are very sensitive to pH value and adjust their activity associated with proton exchange and many others [29], The reactors were evaluated by acidophil (pH < 7), neutral (pH = 7) and alkaline (pH > 7) in most cases. Nearly identical OCV were recorded and acidophil pH reactors showed relatively higher power generation compare to neutral and alkaline [30], Higher anode potential were observed under acidophil condition, while cathode potential were limited to a small range, which indicated the output voltage of an MFC reactor is more related to anode condition [31],
1.3.3 Electric Conductivity and MFCs
The electric conductivity (EC) of the MFC reactor is another main aspect that could have influence on the MFC power generation. In most of the MFC systems, a buffer solution is needed, which can maintain the solution pH value, on order for the bacteria to stay alive [32], When the concentration of a buffer solution increase, the power output, in terms of the peak power density, also increases [33], The increase of the MFC power production is primarily because of the
8


increase of the solution conductivity. As the solution conductivity changed, significant changes in ohmic resistance was overserved, although the increase of the power density is not linear with the conductivity [34],
1.3.4 Dissolved Oxygen and MFCs
The dissolved oxygen (DO) come across the membrane of an MFC reactor and dissolved into solution can result in positive redox condition [10], Under both substrate rich and fuel starvation conditions, temporarily exposure of the reactor to the oxygen did not inhibit the power generation [35], This can happen when the solution need to be changed operating a batch MFC system. After a longer period of oxygen exposure, the bacteria community also didn’t change much and the voltage output recovery quickly after there were no extra oxygen introduced into the MFC system, and the power density remain at the same level. It suggests despite of the oxygen dissolved in the solution, and the MFCs exoelectrogenic capability remains the same [35],
1.4 Objectives
The main challenges for developing an MFC is producing and extracting the energy as much as possible. The current research topics covers the area including the materials of construction, like anode, cathode and membrane, and the topics about searching the high power densities by changing the bio-chemical variables like pH, chemical oxygen demand (COD) , finding the relationship between microbial community and bio-chemical variables. There are some deficiencies about those researches, the higher power density sometimes doesn’t mean higher overall energy output, and most of the those relation- ships are researched by change biochemical conditions and done separately. It is very hard to find an MFC system that designed to find out how electrical variables interact with the bacteria and those bio-chemical variables in terms of energy production and power output performance.
9


1.4.1 Objective I
Develop an MFC system that can interact with different electrical variables that stimulate the MFCs generating more energy. Find the optimal combination of electrical variables that can produce maximum energy and power under the same initial conditions.
1.4.2 Objective II
Find the relations between electrical variables and bio-chemical variables, by changing the electrical variables, and to study how the bio-system in the MFC reactor response to the electrical stimulus.
1.5 Organization of Thesis
This thesis is organized in the chapters that follow into an operation of the MFC, including the design of the reactor and the experiment, different experiments process and corresponding results are discussed.
Chapter II focuses on the development of the MFC reactor, the built and operation of a single chamber air-cathode reactor, the inoculum preparation and the feeding of MFC during the experiment is discussed.
Chapter III focuses on the development of the interface between electrical part and bio-chemical part, including the design of the circuit board and the probes using to measure the bio-chemical variables, the experiment process and analysis method.
Chapter IV addresses the experiments results and discussion.
The summaries of the results and the summaries are presented in Chapter V.
10


CHAPTER II
MFC CONFIGURATION AND OPERATIONS
2.1 Single-chamber Design and Materials of Construction
The MFC reactor used in the experiments consisted of an anode and cathode placed on opposite sides, and between the electrodes are two plastic (Plexiglas) cylindrical chambers 4 cm long by 3 cm in diameter each, both with the empty bed volume of 28 ml. The anode electrode was made of titanium (Ti), and carbon brush was placed on the Ti rod with a length of 4 cm. The cathode was made of carbon paper containing 0.5 mg/cm1 of Platinum (Pt) (10% of Pt/C catalyst, 30% wet-proofing). The cathode electrode was also made of Ti covered with Pt. There are four extra holes on the top of the plastic chamber near the cathode, where the electrochemical probes are placed.
Figure 2.1: MFC Reactor with Sensor Block.
2.2 Inoculum of MFC and Feed
Bacteria present in waste water were used to inoculate the MFCs as waste water bacteria have been shown to be suitable for electricity production [2], Wastewater (primary clarifier effluent)
11


was obtained from a local wastewater treatment plant and used as inoculum. The substrate was sodium acetate with minerals and vitamins in a 50 mM phosphate buffer solution (PBS). The MFC was operating under steady conditions when the voltage output with a load reaches a stable level. After the system became stable, it was then operated using only the 2 g/L acetate.
2.3 MFC Operation
MFC was operated with different external resistances between the anode and cathode. The whole system was operated under the room temperature, keep around 20°C to 25°C. The reactor was operated under batch mode, one batch was consider ended when the reactor voltage keep at low level (less than 20 mv) for more than 2 hours. 120 mg (2 g/L) of sodium acetate with new 60ml PBS solution was replaced to the reactor at the beginning of every batch.
12


CHAPTER III
EXPERIMENT COMPONENTS AND METHODS
3.1 Test System
Figure 3.1: MFC Reactor Test System.
The test system is connected as showed in the Figure 3.1. The MFC reactor with probes was connected to the circuit board, where there is a microcontroller and a converter, also various combinations of resistance, duty ratio and switching frequency were placed. Four electrochemical probes used in the experiments were calibrated and have been sealed in the reactor ever since. The probes were connected to the probe circuit and then connected to the circuit board. All data was sent and recorded by the computer connected to the circuit board. There is also a oscilloscope connected to the circuit board as a monitor.
3.2 MFC Reactor
The MFC reactor used in the experiments was built as described in the previous chapter. The reactor was inoculated as showed in Figure 3.2. In order to keep the initial condition of each batch same, after the reactor reach a steady state voltage output, no more inoculum were introduced into the reactor. Only the solution mixed with 50 mMPBS and 2 g/L sodium acetate were replaced after the pervious batch was done.
13


Figure 3.2: Reactor during Inoculation. 3.3 Probes
Four different bio-chemical testing probes were used during the experiments, pH, EC, DO and Oxidation-Reduction potential (ORP) (Atlas Scientific LLC, USA), as showed in Figure 3.3.
Figure 3.3: Probes in Reactor.
14


Since all the probes don’t need to recalibrate for a long time, all the probes were put into the reactor through holes on the top of the reactor, submerged into solution and sealed from the air using plumber tape. Then the probes are connected to an associated circuit board.
3.3.1 pH Probe
A pH electrode probe is a passive device that detects a current generated from the hydrogen ion activity. The current that is generated from the hydrogen ion activity is the reciprocal of that activity and can be calculated using,
£=£°lj!nafl+, (3.1)
where R the ideal gas constant. T is the temperature in Kelvin. F is the Faraday constant.
3.3.2 Conductivity Probe
A conductivity probe is just two conductors with fixed surface area at fixed distance from each other, which also known as conductivity cell. The cell distance and surface area is quantified as the conductivity cells K constant. The conductivity can be calculated by equation,
conductivity = — K, (3.2)
rs
Where rs is the resistance of the solution. K is the cell constant, which for the probe used in the experiments is 1.0.
3.3.3 Dissolved Oxygen Probe
The galvanic DO sensor is a device that generates a small voltage from 0 mv to 47 mv depending on the oxygen saturation of the High Density Polyethylene (HDPE) sensing
membrane. The probe circuit board will calculate the dissolved oxygen from probe output
15


voltage expressed in unit of mg/L.
3.3.4 Oxidation-Reduction potential Probe
An ORP probe is a passive device that detects a current generated from the oxidation reduction of water, the current is very weak and the probe circuit can pick it up and calculate the ORP value to mv.
3.4 Control Board
Figure 3.4: Control Board.
The control board consist a microcontroller (TMS320F28335, Texas Instruments, TX, USA), a converter and some other components showed in Figure 3.4. The data from the probe circuit
16


were sent to the microcontroller and recorded into computer.
The converter showed in Figure 3.5 was connected to the reactor and load, then sent data to microcontroller, recorded into computer and showed on an oscilloscope.
ci
There are two ways that the resistance were connected to the reactor one is fixed load and the other is connected with hard switching, as showed in Figure 3.6, and Figure 3.7.
17


Figure 3.7: Load with SM’itching. 3.5 Test and Analysis Method
With the same initial condition in reactor, different load conditions were tested batch by batch. Considering the MFC as a constant current source, then the equivalent load can be calculated
by,
Req = (3-3)
Where r is the actual resistance connected to the circuit board and d is duty ratio. Output voltage was recorded by the board and saved into computer, the data sampling interval was every 3 or 4 seconds, and the instantaneous output power was calculated by,
p2
P=7T’ (34)
Keq
Where v is the MFC reactor output voltage and Req is the equivalent external load. During the experiments, some test were conducted under hard switching, means that the voltage reeding is not the actual voltage produced by MFC reactor. As showed in Figure 3.8.
18


• Actual Voltage
• Voltage Reading
co
£
U
U-
s
Figure 3.8: Hard Switching Voltage Reading.
The voltage reading is the average value of MFC open circuit voltage (OCV) and the actual MFC voltage output, so when calculate the MFC output power, if the difference between reading value and actual value was neglected, the v used in the equation should be the reading value from data times the duty ratio d.
The energy was calculated by,
E = f P(t)dt , (3.5)
Where dt is the data recording interval. All the data and figures were analyzed and drafted by Microsoft Excel or MATLAB.
19


CHAPTER IV
EXPERIMEN AND RESULTS
4.1 MFC Connected with Same Equivalent Load
In the experiments, the MFC reactor was connected to a circuit board, where four different resistors (500 Q, 375 Q, 250 Q, 125 Q), three different duty ratio (0.75, 0.5, 0.25), and two different switching frequency (2000 Hz, 10000 Hz) were examined. To maintain a constant current from MFC, and so when the 250 Q resistor was applied with 0.5 duty ratio, 375 LI resistor was applied with 0.75 duty ratio, and 125 Q resistor was applied with 0.25 duty ratio. The average external load should be same as a 500 Q fixed load, so that the MFC reactor were applied the same operating point. Theoretically, these three sets of experiments should produce the same amount of energy under the same feeding condition. The resistance connected to the reactor and duty ratio, switching frequency used in the experiments and the equivalent resistance values are listed in the Table 4.1 below.
Table 4.1: Different Load Condition with Same Operating Point.
Duty Ratio Resistance Connected (Q) Equivalent Resistance (Q)
0.25 125 500
0.50 250 500
0.75 375 500
1.00 500 500
20


4.1.1 Voltage Output during MFC Operation
With the equivalent load connected, showed in Figure 4.1, we can find that except for the fixed 250 Q load, which had the longest extraction time (nearly 140 hours), all the other loads applied with switching frequency and duty ratio had shorter extraction time. It suggested that the electrical stimulation did have impact on the bio-system. The 125 Q load with 0.25 duty ratio showed the highest steady stage voltage, around 350 mv when the switching frequency is
Voltage
500
----SOOohm ----375ofim@duiyO.75 10kHz----375ofim@duiyO.75 2kHz ----250ohm@duty0.510kHz
----250ohin@dutyO,S2kHz ------125otim@dutyO,2S 10kHz----125otim@dutyO.2S 2kHz
Figure 4.1: MFC Voltage under Same Equivalent Load.
10000 Hz and 470 mv when the switching frequency is 2000 Hz. In both batch the extraction time were around 50 hours. The 250 Q load with showed similar value of 330 mv steady stage voltage but with the switching frequency of 10000 Hz the extraction time last twice (nearly 80 hours) as with the switching frequency of 2000 Hz (around 40 hours). The 375 LI load with 0.75 duty ratio showed the extraction time were similar between both 10000 Hz and 2000 Hz
21


(around 40 hours), but the steady stage voltage with 2000 Hz switching frequency is 50 mv
higher than the one with 10000 Hz switching frequency, 300 mv to 250 mv. The steady stage voltage decreases in turns of 125 Q load with 0.25 duty ratio 250 Q load with 0.5 duty ratio 375 Q load with 0.75 duty ratio. Interestingly all the voltage output data showed a small voltage drop during the first hour of each experiment when the load is not fixed 500 Cl. In summary, MFC output voltage wise, the equivalent load with switching and duty showed different pattern compared with fixed load. By applying the switching frequency and duty ratio, the steady stage voltage went higher and the extraction time last shorter, in some cases higher switching frequency indicated higher output voltage, in some cases higher switching frequency indicated longer extraction time but there were no significant trend found between different switching frequency or duty ratio.
4.1.2 Power Output during MFC Operation
Figure 4.2: MFC Power under Same Equivalent Load.
22


Due to the power calculation method using equation 3.4 and the equivalent external loads had
same 500 Q value, so MFC power output of the reactor showed the similar shape as the MFC voltage output, showed in Figure 4.2, but if we consider the fixed load as duty ratio of 1, and then higher duty ratio, higher power output can be observed. Since the less extraction time with load applied duty ratio were found, also lower duty ratio indicate the less time the reactor was actually connected to the load, caused the reactor produced less power.
4.1.3 Energy Output during MFC Operation
Energy


£ 30

1°

< ) 20 40 60 80 100 120 1-40 160 180 Time [Hr) 500ohm 37Sohro@dut*0,7S lOfcHr 37Sohm@duttf).7S 2kHz 2SOohm£>dutyO.S 10kHz 250ohm@dutv0,52Wl2 12Sohm^dulv0.2S 10kHz 12Sohm@dvt>0,2S 2kHz
Figure 4.3: MFC Energy under Same Equivalent Load.
Due to the fact that the same equivalent loads were connected to the reactor, then the similar energy output were expected before the experiments, but the result showed some differences. Calculated the energy production using equation 3.5, compared to the 500 Q fixed load energy output, all the resistance applied with duty ratio and switching frequency produced less than half of the energy, showed in Figure 4.3. From Figure 4.4, showed that when the switching
23


frequency was 2000 Hz, the 375 LI load with 0.75 duty ratio had highest energy output, 250 LI
load with 0.5 duty ratio and 125 Cl load with 0.5 duty ratio had similar lower energy output. From Figure 4.5, showed that when the switching frequency was 10000 Hz, the 250 Q load with 0.5 duty ratio had highest energy output, 375 Q load with 0.75 duty ratio had second
Energy
10 20 30 40 SO 60
Time (Hr)
----375chnt0dutyO. 75 2kHi ------2b0afwn@0$2kHr --------12Sohmpdutv0.25 2kHr
Figure 4.4: MFC Energy under Same Equivalent 2000 Hz Load.
Figure 4.5: MFC Energy under Same Equivalent 10000 Hz Load.
24


highest energy output, 125 Q load with 0.25 duty ratio had lowest energy output. All six tests
produced energy under the range from 13 Jto 23 J. Two tests using 125 LI load with 0.25 duty ratio under 2000 Hz or 10000 Hz switching frequency were both even lower than 13 J energy output. Despite the load applied with duty ratio and switching frequency had higher steady stage voltage but due to much shorter extraction time, less energy outputs were observed. It may result from two reasons, the switching loss or the electric stimulation reduced the ’willingness’ of bacteria to generate electricity. In conclusion, the duty ratio and switching frequency significantly reduced the energy output, but there no clear patterns in terms of the energy output with switching frequency or duty ratio. This may suggest that the bio-community changed after the electrical stimulations, and with different bacteria in the reactor, the energy produced varied by time.
25


4.1.4 Solution pH Value during MFC Operation
Figure 4.6: MFC pH under Same Equivalent Load.
In Figure 4.6, Figure 4.7, Figure 4.8, the pH increase in the first half hour in most of the tests. These spikes in the results may cause by the reasonable amount of food, sodium acetate were added to the reactor at the beginning of each batch. After the first half hour, the pH value dropped down to 7.5, which is close to the pH value of 50 mMPBS.
26


Figure 4.7: MFC pH under Same Equivalent 2000 Hz Load.
The pH value kept growing in the following hours at a similar slop, with some small spikes in
some batches.
Figure 4.8: MFC pH under Same Equivalent 10000 Hz Load
This indicated that the bacteria in the reactor were consuming the organic matters in solution
27


at a similar pace. In summary the pH value are not affected by the electrical load and electrical
stimulations.
4.1.5 Solution Conductivity Value during MFC Operation
The electric conductivity values in solution are mainly determined by the ionic strength [33], So, the same conductivity value at the starting point of each batch were expected, since the same solution recipe were used in the experiments, but the results showed differently (Figure 4.9).
14000 12000 10000 8000
Â¥ o
6000 4000 2000 0
0 20 40 60 80 100 120 140
Time (Hr)
--SOOohm ---375ohm£>dutyQ.7S 10kHz-37Sohn£ido1Y0.7$ 2klU -250ohm^dutv0.510k
2SOohnvgdutyO.S 2kHr 125ohn>gduty0.2S lOkHt 12Soh«ng0,2S 2kHj
Figure 4.9: MFC EC under Same Equivalent Load.
The batch that connected with 125 Q resistance with 0. 25 duty ratio and 10000 Hz switching
frequency had much higher conductivity value at the beginning (Figure 4.11). In the group
with 2000 Hz switching frequency (Figure 4.10), only the batch which the load were 250 LI
resistance with 0. 5 duty ratio had slightly higher conductivity average value. In the group with
10000 Hz switching frequency (Figure 4.10), only the batch which the load were 375 LI
28


resistance with 0. 75 duty ratio had average conductivity value of 8000 s cm, 20% lower compared to other batches conductivity average value (about 10500 s cm). Excluding the data of these two batches, the average value of solution conductivity with 10000 Hz switching frequency is more close to the value recorded during the 500 Q fixed load was connected, and
Electric Conductivity
-SOOohni ----------375ohff)0O5 2kH; ----------125«*in»p>duryO-2S 2kH
Figure 4.10: MFC EC under Same Equivalent 2000 Hz Load.
29


Electric Conductivity
14000
---SOOofwri 37Sohm£KfoitvO 75 lOkHi 25Oohm0O.5 lOcH? — 12Schm#>dutv0.2S lOfcM
Figure 4.11: MFC EC under Same Equivalent 10000 Hz Load. the value were 25% higher (10500 s cm to 800 s cm) than when the resistances were applied
with 2000 Hz switching frequency. It is also corresponding to the reactor voltage output, higher
voltage leaded to higher conductivity value. In all experiments, test results showed that
conductivity value had a trend of growing up alone with the extraction time. Also in about
every 15 hours, there were peak values recorded during each batch. These results may relate
to the bioactivity in the reactor, higher switching frequency often lead to higher conductivity
in the solution, suggest stronger bioactivity and produce higher voltage. In summary applying
higher switching frequency to an MFC load under the same equivalent operating point, leads
to higher conductivity value in solution in most cases.
30


4.1.6 DO Value during MFC Operation
Figure 4.12: MFC DO under Same Equivalent Load.
The dissolved oxygen value showed in Figure 4.12, Figure 4.13, and Figure 4.14 that in every
batch, the value was very high (around 5 mg L) during the first half hour of each batch, and
then after the reactor reach the stable state voltage, the dissolved oxygen value dropped down
31


Figure 4.13: MFC DO under Same Equivalent 2000 Hz Load.
Dissolved Oxygen
Figure 4.14: MFC DO under Same Equivalent 10000 Hz Load. to less than Img/L till the end of the batch. This may cause by the oxygen introduced to the
reactor when the solution had been changed before each batch. The dissolved oxygen value
showed almost identical trend and average value, regardless of how the external loads were
32


applied. So in conclusion, the dissolved oxygen value will not be affected by the external
electrical stimulations.
4.1.7 ORP Value during MFC Operation
Redox Potential
soo
----SOOohm -----J7Sohm^Wutv0.75 lOldtz----37Sohm^duty0.7S 2kHz ----2S0tfwnp ----2SOohmpdutvO.S 2kHz ------12$ohm^dutv0.2S 10kHz-----12Sohm^dury0.2S 2kHz
Figure 4.15: MFC ORP under Same Equivalent Load.
Figure 4.16: MFC ORP under Same Equivalent 2000 Hz Load.
33


The Oxidation-Reduction potential (ORP) value indicate the potential of solution producing
the electrons, so that with the different load conditions, different ORP value were expected before the experiments. In the results in (Figure 4.15, Figure 4.16, and Figure 4.17), the ORP
Redox Potential
— SOOchm ■■■ 37Sohm^dutvdutv0.S ICfcH? 12Sc*wng>dm>0.?S lOfcHt
Figure 4.17: MFC ORP under Same Equivalent 10000 Hz Load. value decreased rapidly in all batches at the beginning, after reached -500 mv, the ORP value
stayed there with some small spikes for almost entire time until it went back to 0 mv at the end
of the batch. Only in the case of 125 Q resistance with 0. 25 duty ratio and 10000 Hz switching
frequency, the ORP value stayed around -400 mv. Excluding that set of data, we can conclude
that the electrical perimeters don’t have influence on the MFCs ORP value and don’t affect the
MFCs electron production ability.
34


4.1.8 Relation between Duty Switching Frequency and Four Biochemical Values
In order to study the influence of combination with both duty ratio and switching frequency on the biochemical parameters, 3D figures were drafted. The average value of pH, conductivity,
DO and ORP were calculated using,
Data Average
Sum of data recorded Total numbers of data recorded
(4.1)
10000
Duty Ratio 0.2 2000
SW
Figure 4.18: Average MFC pH under Same Equivalent Load.
35


10000
Figure 4.19: Average MFC EC under Same Equivalent Load.
Duty Ratio 0.2 2000 gW
Figure 4.20: Average MFC DO under Same Equivalent Load.
36


-300
Figure 4.21: Average MFC ORP under Same Equivalent Load.
As showed in the Figures 4.18 to Figure 4.21, there were no clear trend for all four biochemical
parameters. The average value of those biochemical parameters jumped up and down with
different duty ratios and switching frequencies. In summary, when the MFC connect to the
external resistance that applied with duty ratio and switching frequency, then even with the
equivalent load value, the MFCs output voltage, power and energy are different. Base on the
results, we can only conclude that with the duty ratio and switching frequency applied, the
MFCs energy extraction time will become shorter. For biochemical perimeters, the value of
pH, DO, and ORP will not be affected by the duty ratio and switching frequency. The higher
switching frequency will result in higher conductivity value in the solution.
4.2 MFC Connected with Loads Applied Different Switching Frequency In the experiments, the MFC reactor was connected to a circuit board, where four different resistors (500 Q, 375 Q, 250 Q, 125 Q), two different switching frequency (2000 Hz, 10000 Hz) were applied with same 0.5 duty ratio. The same data recording and analyzing methods
37


were used as described in the previous section. The resistance connected to the reactor and
duty ratio, switching frequency used in the experiments and the equivalent resistance values are listed in the Table 4.2 below.
Table 4.2: Different Load Condition with Same Switching Frequency.
Duty Ratio Resistance Connected(Q) Equivalent Resistance(Q)
0.50 125 250
0.50 250 500
0.50 375 750
0.50 500 1000
4.2.1 Voltage Output during MFC Operation
Voltage output of each batch showed the same shape as in previous batches of experiments (Figure 4.22). Excluding the data from 500 Q with 0.5 duty ratio, the voltage outputs were almost the same, in the set of 250 Q the steady stage voltage reached around 315 mv, in the set of 375 Q the steady stage voltage reached around 350 mv. In each corresponding set, load with 10000 Hz switching frequency, the extraction time last longer than with 2000 Hz.
38


Voltage
600
soo
Figure 4.22: MFC Voltage under Same Load Duty Ratio.
If we consider the fixed load as a switching frequency of infinity, it caused even longer
extraction time according to the pervious result. In conclusion, applying higher switching
frequency to an MFC load, doesn’t affect the steady stage voltage value but result in longer
energy extraction time in most cases.
4.2.2 Power Output during MFC Operation
The power outputs were calculated by the equation 3.4. Excluding the data from 500 Q with 0.5 duty ratio, since almost the same steady stage voltage outputs were recorded, the power outputs were almost the same when the reactor was connected to the same equivalent resistance. The power outputs curves in Figure 4.23 have almost the same shape as the voltage outputs curves’ shape. In conclusion, applying higher switching frequency to an MFC load, doesn’t affect the steady stage voltage value but result in longer power outputs in most cases.
39


Power



T
^ —1 \
j 0 0 0 5 0 G 0 0 0 100
Time |Hr|
-----SOOohmffdutyO.510kHz -------SOOohm^dutyOS 2kHi ---------37Sc*im$PdiityO.S lOkHi — 37Sdtm0dutyO.52km -------------250ohmpdutyO.S lOkH/ --------250o#«n^dut>0 S 2kH*
Figure 4.23: MFC Power under Same Load Duty Ratio.
4.2.3 Energy Output during MFC Operation
Energy
Figure 4.24: MFC Energy under Same Load Duty Ratio.
The energy outputs were calculated by the equation 3.5. As showed in Figure 4.24 and Figure
4.25, excluding the data from 500 Q with 0.5 duty ratio, the reactor produced more energy with
10000 Hz switching frequency than with 2000 Hz switching frequency when the connected to
the corresponding same equivalent load. Since the voltage outputs were almost same at steady
40


stage when connected to the same equivalent load, with 10000 Hz
Energy
25 20 IS
£
I
10 s 0
■ MMohmgKIutvO.S lCkrn ■ 500ohmg>duiy0.5 2kM ■ 37SohmenhJtv0.5 10km ■ 37Sohm^dutyO.S 2kHz ■ 2S0c#iirig>d0.5 10kH2 ■250oh«ngidut>0.5 2kH2
Figure 4.25: MFC Energy Chart under Same Load Duty Ratio. switching frequency the reactor can be extracted for longer time, so the energy outputs with
10000 Hz switching frequency were higher. In conclusion, applying higher switching
frequency to an MFC load, could result in more energy in most cases.
41


EC Is/cm] pH Value
4.2.4 Biochemical Values Output during MFC Operation
pH
12
0 1----
0 10 20 30 40 50 60 70 80 90 100
Time (Hr]
----500Qhm@duTyO.510k.H2 -500ohm@duty0.5 2kH2 -375ohm@duty0.510kH2 -375ohm@duty0.5 2kH2 --250ohm@duty0.510kH2 -2SOohm@dutyO.5 2m2
Figure 4.26: MFC pH under Same Load Duty Ratio.
Electric Conductivity
12000
—500ohm@duty0.5 !0kH2 SOOohmpdutyO.S 2kHz 375ohm@duty0 5 10kH2 375ohm@dutv0.52kHi 250olwn@dutyO 5 10k M2 250ohm@duty0.5 2kH2
Figure 4.27: MFC EC under Same Load Duty Ratio.
42


Dissolved Oxygen
t**l*rmr w •#*'** Vi**** *"*,*,'f.ifM v|i, .y-
kjuuig
SO 60
90 100
Time (Hr)
-SOOolunifBduryOS 10kHz ------SOOohm^duCvO.S 2kMz — 37Sobnn*dury0.S 10kHz — 37Sotim$>dutvO.S2ldtz
m«&dtitvO,S 10kHz — 2S0ohiti«?dirt>0.S 2 kHz
Figure 4.28: MFC DO under Same Load Duty Ratio.
Redox Potential
-500otwn$>dutv0.510kHz -----SOOchm^dulyO S 2kHz -------37SohnH»dutyO.S ICfcHz
OS2kHz ------2SOohmg»dutyO.S10kHz ------2SOcfan£>du1vO.S2kHz
Figure 4.29: MFC ORP under Same Load Duty Ratio.
All four biochemical parameters showed the shame shape as the results from the previous
batches, and for pH, DO, ORP values, there was no clear difference found be- tween
10000 Hz or 2000 Hz switching frequency. With 10000 Hz switching frequency, higher
43


solution conductivity value were showed in the results, which confirmed the result in the
previous section. In summary, by only changing the switching frequency, it doesn’t affect the pH, DO, ORP values in the MFC reactor, and higher switching frequency may indicate higher solution conductivity.
4.3 MFC under Different Operating Points
In the experiments, the MFC reactor was connected to a circuit board, where three different resistors (500 Q, 375 Q, 250 Q), three different duty ratio (0.75, 0.5, 0.25), and all operated under 10000 Hz switching frequency. The resistance connected to the reactor and duty ratio, switching frequency used in the experiments and the equivalent resistance values are listed in the table below.
Table 4.3: Different Load Condition under Different Operating Points.
Duty Ratio Resistance Connected(Q) Equivalent Resistance(Q)
0.25 250 1000
0.25 375 1500
0.25 500 2000
0.50 250 500
0.50 375 750
0.50 500 1000
0.75 250 333
0.75 375 500
0.75 500 666
44


4.3.1 MFC Voltage and Power Output
In order to evaluate the performance of the reactor under different operating points and different load conditions, a polarization test was examined as control group. Polarization curves were obtained by the single-cycle method [36], various external resistances were connected across the MFC, with each resistance being connected for several minutes and the voltage recorded using a circuit board and computer. The polarization curves is in Figure 4.30
and Figure4.31.
700 600 1 500 l— cu 0 1 400 66 J, 300 QJ 00 ro ? 200 > Polarization Curve

^ — 4


, â–  â– 


1UU 0 (
) 0.5 1 1.5 2 2.5 ; Current [mA] S 3.5
Figure 4.30: MFC Polarization Curve.
45


700 600 J. 500 1— CL) O !. 400 o3 £_ 300 100 0 ( Polarization Curve



, ..



) 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Current [mA] L
Figure 4.31: MFC Polarization Curve Part.
In order to evaluate the actual experimental data, the median value of voltage and power were
calculated from the stable stage of each batch of the experiment which can represent the reactor performance and the value were listed in the Table4.4 below.
Table 4.4: Duty ratio, switching frequency and equivalent resistance values.
Vm(777 v) \(mA) P(/nf) Req (Q)
435 0.2175 95 2000
406 0.2707 110 1500
352 0.3520 123 1000
340 0.4533 154 750
324 0.4865 157 666
307 0.6140 189 500
262 0.7868 206 333
The MFC power is calculated by equation3.4, and MFC current is calculated by the equation,
46


(4.2)
I _ vm
~ Req 9
Where Vm is the median value and Req is the equivalent resistance value from the Table4.4.
BOO 450 400 ? s, 350 k_ “ 200 100 50 0 ( Data Curve





“•“V
-*-p



3 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Current [mA]
Figure 4.32: MFC Test Data Curve.
Comparing the Figure 4.31, and Figure 4.32, drafted by data from polarization test and data from the experiment batches, the curve drafted by experimental data showed the same trend as in part of the polarization test, but both voltage and power values were lower compared to the results in polarization test. This suggested that when the resistance were applied with duty ratio and switching frequency, the results from different operating points follow the polarization test, but lower voltage output and lower power output were likely to be produced compared to the equivalent fixed loads.
47


4.3.2 MFC Biochemical Parameters
The median value of four biochemical parameters were calculated from the stable stage of each batch of the experiment which can represent the actual reactor performance and the value were listed in the table below.
Table 4.5: Biochemical Values under Different Load Conditions.
Req (Q) pH EC {s/cm) DO (mg/T) ORP Absolute value(wv)
2000 8.217 9143 0.81 493
1500 8.320 10210 1.07 517
1000 8.870 11690 0.84 538
750 8.350 10010 0.94 499
666 8.046 8878 0.75 489
500 7.844 10470 0.45 512
333 7.934 9452 0.14 425
From Figure 4.33 to Figure 4.33, all biochemical parameters were drafted against the equivalent loads.
48


pH
8.8 8.6 01 -2 8.4 CD X 8.2 Q. 8 7.8 7.6





00
) 500 1000 1500 2000 25 Equivalent Resistance
Figure 4.33: Average MFC pH under Different Equivalent Load.
Electric Conductivity
14000 12000 __ 10000 .§ 8000 “ 6000 ^ 4000
2000 0
0 500 1000 1500 2000 2500
Equivalent Resistance
Figure 4.34: Average MFC EC under Different Equivalent Load.
49


Figure 4.35: Average MFC DO under Different Equivalent Load
Redox Potential
Figure 4.36: Average MFC ORP under Different Equivalent Load.
The figure of biochemical parameter showed certain relations between the external loads. With the changing of operating points, both EC and ORP didn’t change much, EC range from 9000 s cm to 12000 s cm, the ORP value range from -540 mv to -480 mv. Only DO value followed the voltage output, the lower the DO values were, the higher MFC voltage were, and also value showed a linear increase when the equivalent loads were under 750 Cl. It indicates that the
oxygen in the reactor affects the voltage production and DO value may change with the
50


external load when the resistance is relatively low. So we can conclude that there were no linear
relations between external load and the value of solution conductivity ORP, the values were consistent between certain ranges. The solution pH value changes but not related to the external load, with 1000 Q equivalent load, reached the highest value of 8.9. The dissolved oxygen the DO value was inversely proportional to the MFC output voltage.
51


CHAPTER V
CONCLUSION
In conclusion, under the same initial condition when the MFC reactor is connected to the fixed load, the bacteria can produce voltage for a longer time, compared to when connected to the load applied with switching frequency and duty ratio. Also when MFC reactor is connected to the fixed load, it can produce more overall energy. The switching frequency and duty ratio dont have any linear relations with pH, DO, ORP value. These bio-chemical variables mostly depend on the reactors solution condition. The higher switching frequency results in higher EC value and longer MFC electricity producing time in most cases. The pH, EC, ORP values dont affect by changes of equivalent external load values. The DO value has an inversely proportional relations with of equivalent external load values when the load is low.
52


REFERENCES
[1] Richard S Berk and James H Canfield. Bioelectrochemical energy conversion.
Applied microbiology, 12(1): 10—12, 1964.
[2] John B Davis and Harold F Yarbrough. Preliminary experiments on a microbial fuel cell. Science, 137(3530):615-616, 1962.
[3] Komeel Rabaey, Nico Boon, Steven D Siciliano, Marc Verhaege, and Willy Ver- straete. Biofuel cells select for microbial consortia that self-mediate electron transfer. Applied and environmental microbiology, 70(9):5373-5382, 2004.
[4] Daniel R Bond and Derek R Lovley. Electricity production by geobacter sul- furreducens attached to electrodes. Applied and environmental microbiology, 69(3): 1548-1555, 2003.
[5] Yuri A Gorby, Svetlana Yanina, Jeffrey S McLean, Kevin M Rosso, Dianne Moyles, Alice Dohnalkova, Terry J Beveridge, In Seop Chang, Byung Hong Kim, Kyung Shik Kim, et al. Electrically conductive bacterial nanowires produced by shewanella oneidensis strain mr-1 and other microorganisms. Proceedings of the National Academy of Sciences,
103(30): 11358-11363, 2006.
[6] Bruce E Logan. Peer reviewed: extracting hydrogen and electricity from renew- able resources, 2004.
[7] Komeel Rabaey, Nico Boon, Monica H'ofte, and Willy Verstraete. Microbial phenazine production enhances electron transfer in biofuel cells. Environmental science & technology, 39(9):3401-3408, 2005.
53


[8]Shaoan Cheng and Bruce E Logan. Sustainable and efficient biohydrogen produc- tion via
electrohydrogenesis. Proceedings of the National Academy of Sciences, 104(47): 18871-18873, 2007.
[9] Bruce E Logan. Microbial fuel cells. John Wiley & Sons, 2008.
[10] Bruce E Logan, Bert Hamelers, Ren e Rozendal, Uwe Schro' der, Ju'rg Keller, Ste- fano Freguia, Peter Aelterman, Willy Verstraete, and Komeel Rabaey. Microbial fuel cells: methodology and technology. Environmental science & technology, 40(17):5181—5192,
2006.
[11 ]Hong Liu and Bruce E Logan. Electricity generation using an air-cathode single chamber microbial fuel cell in the presence and absence of a proton exchange membrane. Environmental science & technology, 38(14):4040-4046, 2004.
[12] Yanzhen Fan, Hongqiang Hu, and Hong Liu. Enhanced coulombic efficiency and power density of air-cathode microbial fuel cells with an improved cell configu- ration. Journal of Power Sources, 171(2):348-354, 2007.
[13] Eric AZielke. Design of a single chamber microbial fuel cell. Retrieved from Humboldt State Einiversity, School of Engineering Web site: http://www. engr. psu. edu/ce/enve/logan/bioenergy/pdf/Engr 499 final zielke. pdf, 2005.
[14] Daniel R Bond, Dawn E Holmes, Leonard M Tender, and Derek R Lovley. Electrode-reducing microorganisms that harvest energy from marine sediments. Science, 295(5554):483-485, 2002.
54


[15]Komeel Rabaey, Geert Lissens, Steven D Siciliano, and Willy Verstraete. A microbial
fuel cell capable of converting glucose to electricity at high rate and efficiency.
Biotechnology letters, 25(18): 1531-1535, 2003.
[16] Booki Min, Shaoan Cheng, and Bruce E Logan. Electricity generation using membrane and salt bridge microbial fuel cells. Water research, 39(9): 1675-1686, 2005.
[17] SangEun Oh, Booki Min, and Bruce E Logan. Cathode performance as a factor in electricity generation in microbial fuel cells. Environmental science & technology, 38(18):4900-4904, 2004.
[18] Shaoan Cheng, Hong Liu, and Bruce E Logan. Increased power generation in a continuous flow mfc with advective flow through the porous anode and reduced electrode spacing. Environmental science & technology, 40(7):2426-2432, 2006.
[19] Benyi Xiao, Fang Yang, and Junxin Liu. Enhancing simultaneous electricity production and reduction of sewage sludge in two-chamber mfc by aerobic sludge digestion and sludge pretreatments. Journal of hazardous materials, 189(1):444-449, 2011.
[20] Qian Deng, Xinyang Li, Jiane Zuo, Alison Ling, and Bruce E Logan. Power generation using an activated carbon fiber felt cathode in an upflow microbial fuel cell. Journal of Power Sources, 195(4): 1130-1135, 2010.
[21] Ananta Lakshmi Kothapalli. Sediment microbial fuel cell as sustainable power resource. PhD thesis, The University of Wisconsin-Milwaukee, 2013.
[22] Peter Aelterman, Komeel Rabaey, Hai The Pham, Nico Boon, and Willy Ver- straete.
55


Continuous electricity generation at high voltages and currents using stacked microbial fuel
cells. Environmental science & technology, 40(10):3388- 3394, 2006.
[23] Minghua Zhou, Meiling Chi, Jianmei Luo, Huanhuan He, and Tao Jin. An overview of electrode materials in microbial fuel cells. Journal of Power Sources, 196(10):4427-4435, 2011.
[24] Bruce Logan, Shaoan Cheng, Valerie Watson, and Garett Estadt. Graphite fiber brush anodes for increased power production in air-cathode microbial fuel cells. Environmental science & technology, 41(9):3341-3346, 2007.
[25] Sang-Eun Oh and Bruce E Logan. Proton exchange membrane and electrode sur- face areas as factors that affect power generation in microbial fuel cells. Applied microbiology and biotechnology, 70(2): 162-169, 2006.
[26] Cristian Picioreanu, Ian M Head, Krishna P Katuri, Mark CM van Loosdrecht, and Keith Scott. A computational model for biofilm-based microbial fuel cells. Water Research, 41(13):2921-2940, 2007.
[27] Cristian Picioreanu, KP Katuri, IM Head, Mark CM van Loosdrecht, and Keith Scott. Mathematical model for microbial fuel cells with anodic biofilms and anaerobic digestion. Water science and technology, 57(7):965-971, 2008.
[28] Krishna P Katuri, Keith Scott, Ian M Head, Cristian Picioreanu, and Tom P Curtis. Microbial fuel cells meet with external resistance. Bioresource technology, 102(3):2758-2766, 2011.
56


[29]Barry P Rosen and Simon Silver. Ion transport in prokaryotes. Academic Press, 2014.
[30] Eric R Olson. Influence of ph on bacterial gene expression. Molecular microbi- ology, 8(1):5—14, 1993.
[31] S Venkata Mohan, G Mohanakrishna, B Purushotham Reddy, R Saravanan, and PN Sarma. Bioelectricity generation from chemical wastewater treatment in mediatorless (anode) microbial fuel cell (mfc) using selectively enriched hydro- gen producing mixed culture under acidophilic microenvironment. Biochemical Engineering Journal, 39(1): 121— 130, 2008.
[32] Geun-Cheol Gil, In-Seop Chang, Byung Hong Kim, Mia Kim, Jae-Kyung Jang, Hyung Soo Park, and Hyung Joo Kim. Operational parameters affecting the per- formannce of a mediator-less microbial fuel cell. Biosensors and Bioelectronics, 18(4):327-334, 2003.
[33] Zhuwei Du, Haoran Li, and Tingyue Gu. A state of the art review on micro- bial fuel cells: a promising technology for wastewater treatment and bioenergy. Biotechnology advances, 25(5):464-482, 2007.
[34] Joo-Youn Nam, Hyun-Woo Kim, Kyeong-Ho Lim, Hang-Sik Shin, and Bruce E Logan. Variation of power generation at different buffer types and conductivities in single chamber microbial fuel cells. Biosensors and Bioelectronics, 25(5): 1155— 1159, 2010.
[35] SE Oh, JR Kim, J-H Joo, and BE Logan. Effects of applied voltages and dis- solved oxygen on sustained power generation by microbial fuel cells. Water science and technology, 60(5): 1311-1317, 2009.
57


[36]Valerie J Watson and Bruce E Logan. Analysis of polarization methods for elimination
of power overshoot in microbial fuel cells. Electrochemistry Commu- nications, 13(1):54—56, 2011.
58


Full Text

PAGE 1

EFFECTS OF RESISTANCE, FREQUENCY, DUTY RATIO ON MICROBIAL FUEL CELLS (MFC) PERFORMANCE by SHUO FENG BS, Beijing University of Technology, 2014 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in parti al fulfillment of the requirements for the degree of Master of Science Electrical Engineering Program 2017

PAGE 2

ii This thesis for the Master of Science degree by Shuo Feng has been approved for the Department of Electrical Engineering Program by Jaed o Park, Chair Timberley Roane Jung Jae Lee July 12 th , 2017

PAGE 3

iii Feng, Shuo (M.S., Electrical Engineering Program) Effects of Resistance, Frequency, Duty Ratio on Microbial Fuel Cells (MFC) Perfor mance Thesis directed by Associate Professor Jaedo Park ABSTRACT The performance of a single chamber batch feed microbial fuel cell (MFC) was investigated under different external load conditions. The combination of four differ ent resistances (125 , 250 375 , 500 ), four different duty ratios (0.25, 0.50, 0.75, 1.00) and two different switching frequencies (2000 Hz , 10000 Hz ) were tested during the experiment. Four bio chemical variables were also monitored during the exper iment, including pH, electric conductivity (EC), dissolved oxygen (DO), Oxidation Reduction potential (ORP). The relations between MFC voltage, power, and energy outputs and load conditions were discussed, lower power and energy production were found when the loads were not fixed. The relations between MFC four bio chemical variables and load conditions were also discussed. The pH, DO, ORP values changes were not related to the load conditions changes, and longer energy extraction time and higher EC value were found when the switching frequency was higher. No other clear relations were found during the experiment. It shows the electrical stimulations may affect the bio chemical activity inside the MFC reactor, but not in the way of pH, DO and ORP. The form and content of this abstract are approved. I recommend its publication. Approved: Jaedo Park

PAGE 4

iv This file is dedicated to my family, who support me to finish writing my thesis and support me for the Master program.

PAGE 5

v ACKNOWLEDGMENT S This thesis would not ha ve been possible without the generous support of Dr. Park, Dr. Roane and Dr. Lee. Also, I would like to thank you Alaraj Muhannad, Babaiah gari Bhanu and all other members in ECRF group for the help throughout the whole project.

PAGE 6

vi TABLE OF CONTENTS C H APTER I. I N TRODUCTION ................................ ................................ ................................ .. 1 1.1 Microbial Fuel Cells for Electricity Generation ................................ ........... 1 1.2 Microbial Fuel Cells Configurations and Mate rials ................................ ..... 2 1.2.1 Single chamber reactors ................................ ................................ ....... 2 1.2.2 Two chambers reactors ................................ ................................ ........ 3 1.2.3 Upflow, Tubular type MFC (UMFC ................................ .............. 4 1.2.4 Sediments reactor ................................ ................................ ............. 5 1.2.5 MFC Stacks ................................ ................................ ..................... 6 1.2.6 Chamber materials ................................ ................................ ........... 6 1.2.7 Anode materials ................................ ................................ ............... 6 1.2.8 Cathode materials ................................ ................................ ............. 7 1.2.9 Membrane ................................ ................................ ........................ 7 1.3 Current MFC Research Topics ................................ ................................ .......... 7 1.3.1 Load and MFCs ................................ ................................ ............... 7 1.3.2 pH and MFCs ................................ ................................ ................... 8

PAGE 7

vii 1.3.3 Electric Conductivity and MFCs ................................ ..................... 8 1.3.4 Dissolved Oxygen and MFCs ................................ .......................... 9 1.4 Objectives ................................ ................................ ................................ .... 9 1.4.1 Objective I ................................ ................................ ....................... 10 1.4.2 Objective II ................................ ................................ .................. 10 1.5 Organization of Thesis ................................ ................................ ................ 10 II. MFC CO NFIGURSTION AND OPERATIONS ................................ ................... 11 2.1 Single chamber Design and Materials of Construction ................................ 11 2.2 Inoculum of MFC and Feed ................................ ................................ ................. 11 2.3 MFC Operation ................................ ................................ ........................... 12 III. E XPERIMENT COMPONENTS AND METHODS ................................ ............ 13 3.1 Test System ................................ ................................ ................................ . 13 3.2 MFC Reactor ................................ ................................ ............................... 13 3.3 Probes ................................ ................................ ................................ .......... 14 3.3.1 pH Probe ................................ ................................ .......................... 1 5 3.3.2 Conductivity Probe ................................ ................................ .......... 15 3.3.3 Dissolved Oxygen Probe ................................ ................................ . 15 3.3.4 Oxidation Reduction potential Probe ................................ .............. 16

PAGE 8

viii 3.4 Control Board ................................ ................................ ............................. 16 3.5 Test and Analysis Method ................................ ................................ .......... 18 IV. E XPERIMENT AND RESULTS ................................ ................................ ........... 20 4.1 MFC Connected with Same Equivalent Load ................................ ............ 20 4.1.1 Voltage Output during MFC Operation ................................ .......... 2 1 4.1.2 Power Output during MFC Operation ................................ ............. 22 4.1.3 Energy Output during MFC Operation ................................ ........... 22 4.1.4 Solution pH Value during MFC Operat ion ................................ ..... 2 6 4.1.5 Solution Conductivity Value during MFC Operation ..................... 2 8 4.1.6 DO Value during MFC Operation ................................ ................... 3 1 4.1.7 ORP Value during MFC Operation ................................ ................. 3 3 4.1.8 Relati on between Duty Switching Frequency and Four Bio chemical Values ................................ ................................ ................................ ....... 3 5 4.2 MFC Connected with Loads Applied Different Switching Frequency34 4.2.1 Voltage Output during MFC Operation ................................ .......... 3 7 4.2.2 Power Output during M FC Operation ................................ ............. 3 9 4.2.3 Energy Output during MFC Operation ................................ ........... 36 4.2.4 Biochemical Values Output during MFC Operation ....................... 4 0

PAGE 9

ix 4.3 MFC under Different Operating Points ................................ ........................ 4 4 4.3.1 MFC Voltage and Power Output ................................ ..................... 4 5 4.3.2 MFC Biochemical Parameters ................................ ......................... 4 8 V. C ONCLUSION ................................ ................................ ................................ ...... 5 2 R E FERENCE ................................ ................................ ................................ ............... 5 3

PAGE 10

x TABLES T ABLE 4.1 Different Load Condition with Same Operating Point. ................................ .... 20 4.2 Different Load Cond ition with Same Switching Frequency. ............................ 3 8 4.3 Different Load Condition under Different Operating Points. .......................... 4 4 4.4 Duty ratio, switching frequency and equivalent resistance values. .................. 4 6 4.5 Biochemical Values under Different Load Conditions. ................................ .... 4 8

PAGE 11

xi FIGURES F I GURE 1.1 General Oper ating principles of a MFC. ................................ .......................... 2 1.2 Schematic Diagram of Single chamber MFC. ................................ .................. 3 1.3 Schematic Diagram of Two chambers MFC. ................................ ................... 4 1.4 Schematic Diagram of Upflow MFC. ................................ ................................ 5 1.5 Schematic Diagram of Sediments MFC. ................................ ........................... 5 2.1 MFC Reactor with Sensor Block. ................................ ................................ ...... 11 3.1 MFC Reactor Test System. ................................ ................................ ................ 13 3.2 Reac tor during Inoculation. ................................ ................................ .............. 14 3.3 Probes in Reactor. ................................ ................................ ............................. 1 4 3.4 Control Board. ................................ ................................ ................................ ... 16 3.5 Schematic of Converter. ................................ ................................ .................... 17 3.6 Fixed Load. ................................ ................................ ................................ ........ 17 3.7 Load with Switching. ................................ ................................ ......................... 18 3.8 Hard Switching Voltage Reading. ................................ ................................ ..... 19 4.1 MFC Voltage under Same Equivalent Load . ................................ .................... 21 4.2 MFC Power under Same Equivalent Load . ................................ ...................... 22

PAGE 12

xii 4.3 MFC En ergy under Same Equivalent Load . ................................ ..................... 23 4.4 MFC Energy under Same Equivalent 2000 Hz Load . ................................ ...... 2 4 4.5 MFC Energy under Same Equivalent 10000 Hz Load . ................................ .... 24 4.6 MFC pH under Same Equivalent Load . ................................ ............................ 2 6 4.7 MFC pH unde r Same Equivalent 2000 Hz Load . ................................ ............. 2 7 4.8 MFC pH under Same Equivalent 10000 Hz Load . ................................ ........... 2 7 4.9 MFC EC under Same Equivalent Load ................................ ............................. 2 8 4.10 MFC EC under Same Equivalent 2000 Hz Load ................................ .............. 2 9 4.11 MFC EC under Same Equivalent 10000 H z Load ................................ ............ 3 0 4.12 MFC DO under Same Equivalent Load ................................ ............................ 3 1 4.13 MFC DO under Same Equivalent 2000 Hz Load ................................ ............. 3 2 4.14 MFC DO under Same Equivalent 10000 Hz Load ................................ ........... 3 2 4.15 MFC ORP under Same Equival ent Load ................................ .......................... 3 3 4.16 MFC ORP under Same Equivalent 2000 Hz Load ................................ ........... 3 3 4.17 MFC ORP under Same Equivalent 10000 Hz Load ................................ ......... 3 4 4.18 Average MFC pH under Same Equivalent Load ................................ .............. 3 5 4.19 Average MFC EC under Same Equivalent Load ................................ .............. 3 6 4.20 Average MFC DO under Same Equivalent Load ................................ ............. 3 6

PAGE 13

xiii 4.21 Average MFC ORP under Same Equivalent Load ................................ ........... 3 7 4.22 MFC Voltage under Same Load Duty Ratio. ................................ .................... 3 9 4.23 MFC Power under Same Load Duty Ratio. ................................ ...................... 4 0 4.24 MFC Energy under Same Load Duty Ratio. ................................ ..................... 4 0 4.25 MFC Energy Chart under Same Load Duty Ratio. ................................ ........... 4 1 4.26 MFC pH under Same Load Duty Ratio. ................................ ............................ 4 2 4.27 MFC EC under S ame Load Duty Ratio. ................................ ........................... 4 2 4.28 MFC DO under Same Load Duty Ratio. ................................ ........................... 4 3 4.29 MFC ORP under Same Load Duty Ratio. ................................ ......................... 4 3 4.30 MFC Polarization Curve. ................................ ................................ .................. 4 5 4.31 MFC Polarization Curve Part. ................................ ................................ .......... 4 6 4.32 MFC Test Data Curve. ................................ ................................ ...................... 4 7 4.33 Average MFC pH under Different Equivalent Load. ................................ ........ 4 9 4.34 Average MFC EC under Different Equivalent Load ................................ ........ 4 9 4.35 Average MFC DO under Different Equivalent Load ................................ ........ 5 0 4.36 Average MFC ORP under Different Equivalent Load ................................ ...... 5 0

PAGE 14

1 CHAPTER I INTRODUCTION 1.1 Microbial Fuel Cells for Electricity Generation Microbial fuel cell (MFC) is a rapidly growing research topic in recent decades for its electricity generation capability. An MFC reactor is a bio electro chemical system and an Oxidation reduction reaction is happening in the reactor that use bacteria as the bio catalysts [1][2]. It is a complex system involving the knowledge from different scientific and engineering fields and to analyze the performance of an MFC system as an energy source, the knowledge from microbiology, electro chemistry and electrical engineering are required. In the MFCs systems, bacteria produce electrons by consuming the substrates existing in solution and transfer them to the anode (negative terminal) and electrons can flow to the cathode (positive terminal) through a load, such as a resistor that is connected outside the reactor [2]. The current that generated by the MFC reactor is flowing from cathode to anode which is opposite to the electron flow. There are three ways that electrons can be transferred to the anode discovered. Fi rst is by electron mediators or shuttles [3], second is by direct membrane associated electron transfer [4], and the third way is by so called nanowires [5] produced by the bacteria, even though the mechanism of electron transfer is not well known [6]. The re are various microbial cultures that are known to generate electricity in the MFCs. The research shows that in general MFCs using mixed cultures usually achieve substantially higher power densities than those with pure cultures [7]. Theoretically any bio degradable organic matter can be used to build an MFC reactor, including acids, alcohols, carbohydrates, proteins, and cellulose [8][9]. Various oxidants for the reduction reaction at cathode (electron acceptor) have been used in MFCs, such as oxygen, nitr ate, ferricyanide, permanganate, ferric iron [10]. Wastewater and oxy gen are the most promising

PAGE 15

2 electron donor and acceptor, respectively, as they are the most accessible and economical [9]. In typical air cathode MFC reactors the electrons will combine at the cathode with the protons that come from the anode and oxygen provided from air; so the side product is just water. 1.2 Microbial Fuel Cells Configurations and Materials MFCs are being constructed using a variety of materials, and in an increasing diversity of configurations. The most common designs include the single chamber air cathode reactor, two chamber reactor used in the research laboratory and sediments reactor used in the field. The general operating principle of a MFC reactor is showed in Figure 1.1 [10], and some of the MFC designs are discussed in this section. Figure 1.1: General Operating principles of a MFC. 1.2.1 Single chamber R eactors Consisting only the anode chamber, in this configuration, the anode and cathode are placed on eithe r side of the chamber, usually cube shape. The cathode is exposed to air on one side and the solution on the other side, inside of the anode chamber. The proton exchange membrane (PEM) used in this system primarily to keep water from leaking through the ca thode, and it also reduces

PAGE 16

3 oxygen diffusion into the anode chamber. The anode side, usually a brush where the bacteria Figure 1.2: Schematic Diagram of Single chamber MFC. grow, is also sealed in the solution, so that air cannot enter. The single chamber, air cathode designs reactors have higher power density output relative to two chamber reactors, due to its low internal resistance [11]. One of the great challenges for the PEM less MFC is that the Coulombic efficiency is much lower than th ose with PEM [12]. A schematic diagram of single chamber MFC reactor [13] is showed in Figure 1.2. 1.2.2 Two chambers Reactors Consisting two containers, usually two bottles connected by tube, or two cubes connected side by side, containing a separator in the middle, which is usually a PEM such as Nafion or Ultrex, or a plain salt bridge [14] [15] [16]. The important point to this design is using a membrane that allows only protons, not allowing the substrate or electron to pass between the chambers. The am ount of power that generating from these systems is affected by the surface area of the cathode relative to that of the anode [17]. The membrane and longer distance between anode and cathode cause a higher internal resistance within a two chamber MFC react or [18], which result in lower

PAGE 17

4 power density output compare to single chamber reactors. A schematic diagram of two chamber MFC reactor [19] is showed in Figure 1.3 Figure 1.3: Schematic Diagram of Two chambers MFC. 1.2.3 Upflow, Tubular type MFC (UMFC) T here are mainly two kinds of upflow, tubular type MFC reactors. One is just like a two chambers reactor, the anode chamber is put under the cathode chamber, separated by a PEM membrane. The other kind is comprised of a PEM membrane, which was folded and se aled through soldering to provide a cylindrical structure. Two glass stoppers, each with two inlets/outlets were inserted on both sides of this cylinder [20]. The advantage of this design is that it can reduce the anode cathode electrode spacing. A schemat ic diagram of upflow MFC reactor [20] is showed in Figure 1.4.

PAGE 18

5 Figure 1.4: Schematic Diagram of Upflow MFC. 1.2.4 Sediments Reactor It is one of the MFC applications, using carbon cloth as anode, which was buried below and parallel to the surface of the sediment. The cathode was positioned in the seawater, above and parallel to the sediment surface. Typically, the seawater layer above the cathode was 5 cm in depth [21]. No separator was used in the cell. So this is a membrane less system. The anode and cathode were maintained as horizontal positions. A schematic diagram of Sediments MFC reactor [21] is showed in Figure 1.5. Figure 1.5 : Schematic Diagram of Sediments MFC.

PAGE 19

6 1.2.5 MFC Stacks Any of those reactor configurations can be used to build an MFC stack. In most cases, the MFC stack was constructed on the basis of tubular air cathode MFCs. Usually individual cells were electrically connected in series by wires and hydraulically joined by conductive substrate flow, the performance degradation phenomenon was observed [22]. The open circuit voltage (OCV) and current flow of stacked MFC were also lower [22]. 1.2.6 Chamber Materials The materials used to build up the reactor chamber must be non toxic to the bacteria. Polyvinylchloride (PVC) plastic tube and Plexiglas are two common materials since they are easy to find, easy to make to any shape and at a low cost. 1.2.7 Anode M aterials An odic materials are required to be conductive, biocompatible, and chemically stable in the reactor solution, but copper is not useful due to the copper ions are toxic to bacteria [10]. Since carbon materials are well suited for bacterial growth and have hig h conductivity, so most of the anode materials are carbon based, including carbon paper, carbon cloth, reticulated vitreous carbon (RVC), graphite fiber brush [23]. Carbon paper and carbon cloth are the most flexible materials among those but very expensiv e. The RVC has been ideal for the continuous up flow MFCs because it has excellent conductivity. The increasing accessible surface area increased current density, so the graphite fiber brush was developed for the MFCs [23]. Graphite brush electrodes can ha ve an extremely high surface area by just having a small diameter [24]. During the experiments, we also found out the plastic brush with the iron rod after an acid reaction showed good performance for the bacteria.

PAGE 20

7 1.2.8 Cathode Materials The same materials that have been described for anodes have also been used as cathodes, and different oxidants have been used as the electron acceptor in MFCs. Ferricyanide ( K 3 [Fe(CN ) 6 ] ) is very popular as an experimental electron acceptor in microbial fuel cells [25], and oxygen is the most suitable electron acceptor for the MFC since its high and oxidation potential and it is everywhere and the free of cost [10]. 1.2.9 Membrane The membrane in MFC reactor designs is often used as the separation of the anode and t he cathode compartments, for example a PEM. The most commonly used membranes are made by Nafion. The use of membranes in MFCs increases internal resistance that decreases power production [18]. Alternatives to Nafion, such as Ultrex CMI also are well suite d for MFC reactor designs at a lower cost [10]. 1.3 Current MFC Research Topics MFC performance can be affected by many different aspects, including the factors like circuit load, internal resistance, solution pH value, temperature and others. In the view point of the electrical engineering, the performance of a MFC reactor is evaluated by its power generating capability, so the literature researches were mainly focused on MFC power output different conditions. 1.3.1 Load and MFCs Differences in the externa l load (resistance) were shown to lead to some significant differences in MFC cell operation. The relation between different external load and MFC reactor current generation and power performance were investigated. Most of those tests were conducted after connected different reactors with same configuration to different load for a long period of time,

PAGE 21

8 recorded data hourly. It is considered that with different fixed resistance load, the bacterial community around the anode are formed differently, higher exte rnal load lead to higher biomass growth, lower current generating [26]. Also the increased external load can cause less electroactive bacteria growth, which result in lower voltage output [27]. Although different bacteria community were formed, no signific ant power output differences were found, according to the power curve. The maximum power density points were similar, even the inoculum preparations were different [28]. 1.3.2 pH and MFCs The internal pH environment of an MFC reactor is one of the important factor for MFC power output, which can influence substrate metabolic activity and impact e and H + generation process, due to that most bacteria are very sensitive to pH value and adjust their activity associated with proton exchange and many oth ers [29]. The reactors were evaluated by acidophil (pH < 7), neutral (pH = 7) and alkaline (pH > 7) in most cases. Nearly identical OCV were recorded and acidophil pH reactors showed relatively higher power generation compare to neutral and alkaline [30]. Higher anode potential were observed under acidophil condition, while cathode potential were limited to a small range, which indicated the output voltage of an MFC reactor is more related to anode condition [31]. 1.3.3 Electric Conductivity and MFCs The el ectric conductivity (EC) of the MFC reactor is another main aspect that could have influence on the MFC power generation. In most of the MFC systems, a buffer solution is needed, which can maintain the solution pH value, on order for the bacteria to stay a live [32]. When the concentration of a buffer solution increase, the power output, in terms of the peak power density, also increases [33]. The increase of the MFC power production is primarily because of the

PAGE 22

9 increase of the solution conductivity. As the s olution conductivity changed, significant changes in ohmic resistance was overserved, although the increase of the power density is not linear with the conductivity [34]. 1.3.4 Dissolved Oxygen and MFCs The dissolved oxygen (DO) come across the membrane of an MFC reactor and dissolved into solution can result in positive redox condition [10]. Under both substrate rich and fuel starvation conditions, temporarily exposure of the reactor to the oxygen did not inhibit the power generation [35]. This can happen when the solution need to be changed operating a batch MFC system. After voltage output recovery quickly after there were no extra oxygen introduced into the MFC sys tem, and the power density remain at the same level. It suggests despite of the oxygen dissolved in the solution, and the MFCs exoelectrogenic capability remains the same [35]. 1.4 Objectives The main challenges for developing an MFC is producing and extra cting the energy as much as possible. The current research topics covers the area including the materials of construction, like anode, cathode and membrane, and the topics about searching the high power densities by changing the bio chemical variables like pH, c hemical o xygen demand ( COD ) , finding the relationship between microbial community and bio chemical variables. There are some overall energy output, and most of the those relation ships are researched by change biochemical conditions and done separately. It is very hard to find an MFC system that designed to find out how electrical variables interact with the bacteria and those bio chemical variables in terms of energy production and power output performance.

PAGE 23

10 1.4.1 Objective I Develop an MFC system that can interact with different electrical variables that stimulate the MFCs generating more energy. Find the optimal combination of electrical variables that can p roduce maximum energy and power under the same initial conditions. 1.4.2 Objective II Find the relations between electrical variables and bio chemical variables, by changing the electrical variables, and to study how the bio system in the MFC reactor respo nse to the electrical stimulus. 1.5 Organization of Thesis This thesis is organized in the chapters that follow into an operation of the MFC, including the design of the reactor and the experiment, different experiments process and corresponding results ar e discussed. Chapter II focuses on the development of the MFC reactor, the built and operation of a single chamber air cathode reactor, the inoculum preparation and the feeding of MFC during the experiment is discussed. Chapter III focuses on the development of the interface between electrical part and bio chemical part, including the design of the circuit board and the probes using to measure the bio chemical variables, the experiment process and analysis method. Chapter IV addresses the experimen ts results and discussion. The summaries of the results and the summaries are presented in Chapter V.

PAGE 24

11 CHAPTER II MFC CONFIGURATION AND OPERATIONS 2.1 Single chamber Design and Materials of Construction The MFC reactor used in the experiments consisted of an anode and cathode placed on opposite sides, and between the electrodes are two plastic (Plexiglas) cylindrical chambers 4 cm long by 3 cm in diameter each, both with the empty bed volume of 28 ml . The anode electrode was made of titanium (Ti), and carbon brush was placed on the Ti rod with a length of 4 cm . The cathode was made of carbon paper containing 0.5 mg/cm 2 of Platinum (Pt) (10% of Pt/C catalyst, 30% wet proofing). The cathode electrode was als o made of Ti covered with Pt . There are four extra holes on the top of the plastic chamber near the cathode, where the electrochemical probes are placed. Figure 2.1: MFC Reactor with Sensor Block. 2.2 Inoculum of MFC and Feed Bacteria present in waste water were used to inoculate the MFCs as waste water bacteria have been shown to be suitable for electricity production [2]. Wastewater (primary clarifier effluent)

PAGE 25

12 was obtained from a local wastewater treatment plant and used as inoculum. The substrate was sodium acetate with minerals and vitamins in a 50 mM phosphate buffer solution (PBS). The MFC was operating under steady conditions when the voltage output with a load reaches a stable level. After the system became stable, it w as then operated using only the 2 g/L acetate . 2.3 MFC Operation MFC was operated with different external resistances between the anode and cathode. The whole system was operated under the room temperature, keep around 20 C to 25 C . The reactor was operate d under batch mode, one batch was consider ended when the reactor voltage keep at low level (less than 20 mv ) for more than 2 hours. 120 mg (2 g/L ) of sodium acetate with new 60 ml PBS solution was replaced to the reactor at the beginning of every batch.

PAGE 26

13 CHAPTER III E XPERIMENT COMPONENTS AND METHODS 3.1 Test System Figure 3.1: MFC Reactor Test System. The test system is connected as showed in the Figure 3.1. The MFC reactor with probes was connected to the circuit board, where there is a microcontroller and a converter, also various combinations of resistance, duty ratio and switching frequency were placed. Four electrochemical probes used in the experiments were calibrated and have been sealed in the reactor ever since. The probes were co nnected to the probe circuit and then connected to the circuit board. All data was sent and recorded by the computer connected to the circuit board. There is also a oscilloscope connected to the circuit board as a monitor. 3.2 MFC Reactor The MFC reactor used in the experiments was built as described in the previous chapter. The reactor was inoculated as showed in Figure 3.2. In order to keep the initial condition of each batch same, after the reactor reach a steady state voltage output, no more inoculum were introduced into the reactor. Only the solution mixed with 50 mM PBS and 2 g/L sodium acetate were replaced after the pervious batch was done.

PAGE 27

14 Figure 3.2: Reactor during Inoculation. 3.3 Probes Four different bio chemical testing probe s were used during the experiments, pH, EC, DO and Oxidation Reduction potential (ORP) (Atlas Scientific LLC, USA), as showed in Figure 3.3. Figure 3.3: Probes in Reactor.

PAGE 28

15 Since all the probes don t need to recalibrate for a long time, all the probes were put into the reactor through holes on the top of the reactor, submerged into solution and sealed from the air using plumber tape. Then the probes are connected to an associated circuit board. 3.3.1 pH Probe A pH electrode probe is a passive device that detects a current generated from the hydrogen ion activity. The current that is generated from the hydrogen ion activity is the reciprocal of that activity and can be calculated using, (3. 1 ) where R the ideal gas constant. T is the temperature in Kelvin. F is the Faraday constant. 3.3.2 Conductivity P robe A conductivity probe is just two conductors with fixed surface area at fixed distance from each other, which also known as conductivity cell. The cell distance and surface area is quantified as the conductivity cells K constant. The conductivity can be calculated by equation, (3.2) Where r s is the resistance of the solution. K is the cell constant, which for the probe used in the experiments is 1.0. 3.3.3 Dissolved Oxygen Probe The galvanic DO sensor is a device that generates a small voltage from 0 mv to 47 mv depending on the oxygen saturat ion of the High Density Polyethylene (HDPE) sensing membrane. The probe circuit board will calculate the dissolved oxygen from probe output

PAGE 29

16 voltage expressed in unit of mg/L . 3.3.4 Oxidation Reduction potential Probe An O R P probe is a passive device that d etects a current generated from the oxidation reduction of water, the current is very weak and the probe circuit can pick it up and calculate the ORP value to mv . 3.4 Control Board Figure 3.4 : Control Board. The control board consist a microcontroller (TMS320F28335, Texas Instruments, TX, USA), a converter and some other components showed in Figure 3.4. The data from the probe circuit

PAGE 30

17 were sent to the microcontroller and recorded into computer. The converter s howed in Figure 3.5 was connected to the reactor and load, then sent data to microcontroller, recorded into computer and showed on an oscilloscope. Figure 3.5: Schematic of Converter. There are two ways that the resistance were connected to the reactor one is fixed load and the other is connected with hard switching, as showed in Figure 3.6, and Figure 3.7. Figure 3.6: Fixed Load.

PAGE 31

18 Figure 3.7: Load with Switching. 3.5 Test and Analysis Method With the same initial condition in reactor, different load conditions were tested batch by batch. Considering the MFC as a constant current source, then the equivalent load can be calculated by, , (3.3) Where r is the actual resistance connected to the circuit board and d is duty ratio. Output voltage was recorded by the board and saved into computer, the data sampling interval was every 3 or 4 seconds, and the instantaneous output power was calculated by , , (3. 4 ) Where v is the MFC reactor output voltage and R eq is the equivalent external load. During the experiments, some test were conducted under hard switching, means that the voltage reeding is not the actual voltage produced by MFC reactor. As showed in Figure 3.8.

PAGE 32

19 Figure 3.8: Hard Switching Voltage Reading . The voltage reading is the average value of MFC open circuit voltage (OCV) and the actual MFC voltage output, so when calculate the MFC output power, if the difference between reading value and actual value was neglected, the v used in the equation shoul d be the reading value from data times the duty ratio d . The energy was calculated by , , (3.5) Where dt is the data recording interval. All the data and figures were analyzed and drafted by Microsoft Excel or MATLAB.

PAGE 33

20 CHAPTER IV EXPERIMEN AND RESULTS 4.1 MFC Connected with Same Equivalent Load In the experiments, the MFC reactor was connected to a circuit board, where four different resistors (500 , 375 , 250 , 125 ), three different duty ratio (0.75, 0.5, 0.25), and two different switc hing frequency (2000 Hz , 10000 Hz ) were examined. To maintain a constant current from MFC, and so when the 250 resistor was applied with 0.5 duty ratio, 375 resistor was applied with 0.75 duty ratio, and 125 resistor was applied with 0.25 duty ratio. The average external load should be same as a 500 fixed load, so that the MFC reactor were applied the same operating point. Theoretically, these three sets of experiments should produce the same amount of energy under the same feeding condition. The re sistance connected to the reactor and duty ratio, switching frequency used in the experiments and the equivalent resistance values are listed in the Table 4.1 below. Table 4.1 : Different Load Condition with Same Operating Point . Duty Ratio Resistance Connected ( ) Equivalent Resistance ( ) 0.25 125 500 0.50 250 500 0.75 375 500 1.00 500 500

PAGE 34

21 4.1.1 Voltage Output during MFC Operation With the equivalent load connected, showed in Figure 4.1, we can find that except for the fixed 250 load, which had the longest extraction time (nearly 140 hours), all the other loads applied with switching frequency and duty ratio had shorter extraction time. It suggested that the electrical stimulation did have impact on the bio with 0.25 duty ratio showed the highest steady stage voltage, around 350 mv when the switching frequency is Figure 4.1: MFC Voltage under Same Equivalent Load . 10000 Hz and 470 mv when the switching frequency is 2000 Hz . In both batch the extraction time were around 50 hours. The 250 load with showed similar value of 330 mv steady stage voltage but with the switching frequency of 10000 Hz the extraction time last twice (nearly 80 hours) as with the switching frequenc y of 2000 Hz (around 40 hours). The 375 load with 0.75 duty ratio showed the extraction time were similar between both 10000 Hz and 2000 Hz

PAGE 35

22 (around 40 hours), but the steady stage voltage with 2000 Hz switching frequency is 50 mv higher than the one with 1 0000 Hz switching frequency, 300 mv to 250 mv . The steady stage voltage decreases in turns of 125 load with 0.25 duty ratio 250 load with 0.5 duty ratio 375 load with 0.75 duty ratio. Interestingly all the voltage output data showed a small voltage drop during the first hour of each experiment when the load is not fixed 500 . In summary, MFC output voltage wise, the equivalent load with switching and duty showed different pattern compa red with fixed load. By applying the switching frequency and duty ratio, the steady stage voltage went higher and the extraction time last shorter, in some cases higher switching frequency indicated higher output voltage, in some cases higher switching fre quency indicated longer extraction time but there were no significant trend found between differ ent switching frequency or duty ratio. 4.1.2 Power Output during MFC Operation Figure 4.2: MFC Power under Same Equivalent Load .

PAGE 36

23 Due to the power calculation method using e quation 3.4 and the equivalent external loads had same 500 value, so MFC power output of the reactor showed the similar shape as the MFC voltage output, showed in Figure 4.2, but if we consider the fixed load as duty ratio of 1, and then higher duty ratio, higher power output can be observed. Since the less extraction time with load applied duty ratio were found, also lower duty ratio indicate the less time the reactor was actually connected to the load, caused t he reactor produced less power. 4.1.3 Energy Output during MFC Operation Figure 4.3: MFC Energy under Same Equivalent Load . Due to the fact that the same equivalent loads were connected to the reactor, then the similar energy output were expected before the experiments, but the result showed some differences. Calculated the energy production using equation 3.5, compared to the 500 fixed load energy output, all the resistance applied with duty ratio and switching frequency produced less than half of the energy, sh owed in Figure 4.3 . From Figure 4.4, showed that when the switching

PAGE 37

24 frequency was 2000 Hz , the 375 load with 0.75 duty ratio had highest energy output, 250 load with 0.5 duty ratio and 125 load with 0.5 duty ratio ha d similar lower energy o utput. From Figure 4.5, showed that when the switching frequency was 10000 Hz 0.75 duty ratio had second Figure 4.4: MFC Energy under Same Equivalent 2000 Hz Load . Figure 4.5: MFC Energy under Same Equivalent 10000 Hz Load .

PAGE 38

25 highest energy output, 125 load with 0.25 duty ratio had lowest energy output. All six tests produced energy under the range from 13 J to 23 J . Two tests using 125 load with 0.25 duty ratio under 2000 Hz or 10000 Hz switching frequency were both even lower than 13 J energy output. Despite the load applied with duty ratio and switching frequency had higher steady stage voltage but due to much shorter extraction time, less energy outputs were observed. It may result from two reasons, the switching loss or the electric stimulation reduced the duty ratio and switching frequency significantly reduced th e energy output, but there no clear patterns in terms of the energy output with switching frequency or duty ratio. This may suggest that the bio community changed after the electrical stimulations, and with different bacteria in the reactor, the energy pro duced varied by time.

PAGE 39

26 4.1.4 Solution pH Value during MFC Operation Figure 4.6: MFC pH under Same Equivalent Load . In Figure 4.6, Figure 4.7, Figure 4.8, the pH increase in the first half hour in most of the tests. These spikes in the results may cause by the reasonable amount of food, sodium acetate were added to the reactor at the beginning of each batch. After the first half hour, the pH value dropped down to 7.5, which is close to the pH value of 50 mM PBS.

PAGE 40

27 Figure 4.7: MFC pH under Same Equi valent 2000 Hz Load . The pH value kept growing in the following hours at a similar slop, with some small spikes in some batches. Figure 4.8: MFC pH under Same Equivalent 10000 Hz Load This indicated that the bacteria in the reactor were consuming the org anic matters in solution

PAGE 41

28 at a similar pace. In summary the pH value are not affected by the electrical load and electrical stimulations. 4.1.5 Solution Conductivity Value during MFC Operation The electric conductivity values in solution are mainly determined by the ionic strength [33]. So, the same conductivity value at the starting point of each batch were expected, since the same solution recipe were used in the experiments, but the results showed differently (Figure 4.9). Figure 4.9: MFC EC und er Same Equivalent Load. The batch that connected with 125 resistance with 0. 25 duty ratio and 10000 Hz switching frequency had much higher conductivity value at the beginning (Figure 4.11). In the group with 2000 Hz switching frequency (Figure 4.10), o nly the batch which the load were 250 resistance with 0. 5 duty ratio had slightly higher conductivity average value. In the group with 10000 Hz switching frequency (Figure 4.10), only the batch which the load were 375

PAGE 42

29 resistance with 0. 75 duty ratio had average conductivity value of 8000 s/cm , 20% lower compared to other batches conductivity average value (about 10500 s/cm ). Excluding the data of these two batches, the average value of solution conductivity with 10000 Hz switching load was connected, and Figure 4.10: MFC EC under Same Equivalent 2000 Hz Load .

PAGE 43

30 Figure 4.11: MFC EC under Same Equivalent 10000 Hz Load . the value were 25% higher (10500 s/cm to 800 s/cm ) than when the resistances were applied with 2000 Hz switching frequency. It is also corresponding to the reactor voltage output, higher voltage leaded to higher conductivity value. In all experiments, test results showed that conductivity value had a trend of growing up alone with the extraction time. Also in about every 15 hours, there were peak values recorded during each batch. These results may relate to the bioactivity in the reactor, higher switching frequency often lead to higher co nductivity in the solution, suggest stronger bioactivity and produce higher voltage. In summary applying higher switching frequency to an MFC load under the same equivalent operating point, leads to higher conductivity value in solution in most cases.

PAGE 44

31 4. 1.6 DO Value during MFC Operation Figure 4.12: MFC DO under Same Equivalent Load . The dissolved oxygen value showed in Figure 4.12, Figure 4.13, and Figure 4.14 that in every batch, the value was very high (around 5 mg/L ) during the first half hour of ea ch batch, and then after the reactor reach the stable state voltage, the dissolved oxygen value dropped down

PAGE 45

32 Figure 4.13: MFC DO under Same Equivalent 2000 Hz Load . Figure 4.14: MFC DO under Same Equivalent 10000 Hz Load . to less than 1 mg/L till the end of the batch . This may cause by the oxygen introduced to the reactor when the solution had been changed before each batch. The dissolved oxygen value showed almost identical trend and average value, regardless of how the external loads were

PAGE 46

33 applied. S o in conclusion, the dissolved oxygen value will not be affected by the external electrical stimulations. 4.1.7 ORP Value during MFC Operation Figure 4.15: MFC ORP under Same Equivalent Load . Figure 4.16: MFC ORP under Same Equivalent 2000 Hz Load .

PAGE 47

34 The Oxidation Reduction potential (ORP) value indicate the potential of solution producing the electrons, so that with the different load conditi ons, different ORP value were expected before the experiments. In the results in (Figure 4.15, Figure 4.16, and Fi gure 4.17), the ORP Figure 4.17: MFC ORP under Same Equivalent 10000 Hz Load . value decreased rapidly in all batches at the beginning, after reached 500 mv , the ORP value stayed there with some small spikes for almost entire time until it went back to 0 mv at the end of the batch. Only in the case of 125 resistance with 0. 25 duty ratio and 10000 Hz switching frequency, the ORP value stayed around 400 mv . Excluding that set of data, we can conclude that the electrical perimeters don t have influence on the MFCs ORP value and don MFCs electron production ability.

PAGE 48

35 4.1.8 Relation between Duty Switching Frequency and Four Biochemical Values In order to study the influence of combination with both duty ratio and switching frequency on the bi ochemical parameters, 3D figures were drafted. The average value of pH, conductivity, DO and ORP were calculated using , , (4.1) Figure 4.18: Average MFC pH under Same Equivalent Load .

PAGE 49

36 Figure 4.19: Average MFC EC under Same Equivalent Load . Figure 4.20: Average MFC DO under Same Equivalent Load .

PAGE 50

37 Figure 4.21: Average MFC ORP under Same Equivalent Load . As showed in the Figures 4.18 to Figure 4.21, there were no clear trend for all four biochemical parameters. The average value of those biochemical parameters jumped up and down with different duty ratios and switching frequencies. In summary, when the MFC connect to the e xternal resistance that applied with duty ratio and switching frequency, then even with the equivalent load value, the MFCs output voltage, power and energy are different. Base on the results, we can only conclude that with the duty ratio and switching fre quency applied, the MFCs energy extraction time will become shorter. For biochemical perimeters, the value of pH, DO, and ORP will not be affected by the duty ratio and switching frequency. The higher switching frequency will result in higher conductivity value in the solution. 4.2 MFC Connected with Loads Applied Different Switching Frequency In the experiments, the MFC reactor was connected to a circuit board, where four different resistors (500 , 375 , 250 , 125 ) , two different switching frequency (2000 Hz , 10000 Hz ) were applied with same 0.5 duty ratio. The same data recording and analyzing methods

PAGE 51

38 were used as described in the previous section. The resistance connected to the reactor and duty ratio, switching frequency used in the experiments and the equivalent resistance values are listed in the Table 4.2 below. Table 4.2: Different Load Condition with Same Switching Frequency . Duty Ratio 0.50 125 250 0.50 250 500 0.50 375 750 0.50 500 1000 4.2.1 Voltage Output during MFC Operation Voltage output of each batch showed the same shape as in previous batches of experiments (Figure 4.22). Excluding the data from 500 with 0.5 duty ratio, the voltage outputs were almost the same, in the set of 250 the steady stage voltage reached around 315 mv , in the set mv . In each corresponding set, load with 10000 Hz switching frequency, the extraction time last longer than with 2000 Hz .

PAGE 52

39 Figure 4.22: MFC Voltage under Same Load Duty Ratio. If we consider the fixed load as a switching frequency of infinity, it caused even longer extraction time according to the pervious result . In conclusion, applying higher switching energy extraction time in most cases. 4.2.2 Power Output during MFC Operation The power outputs were calculated by the equation 3.4. Excluding the data from 500 with 0.5 duty ratio, since almost the same steady stage voltage outputs were recorded, the power outputs were almost the same when the reactor was connected to the same eq uivalent resistance. The power outputs curves in Figure 4.23 have almost the same shape as the voltage outputs affect the steady stage voltage value but result in lon ger power outputs in most cases.

PAGE 53

40 Figure 4.23: MFC Power under Same Load Duty Ratio . 4.2.3 Energy Output during MFC Operation Figure 4.24: MFC Energy under Same Load Duty Ratio. The energy outputs were c alculated by the equation 3.5. As showed in Figure 4.24 and Figure 4.25, excluding the data from 500 with 0.5 duty ratio, the reactor produced more energy with 10000 Hz switching fre quency than with 2000 Hz switch ing frequency when the connected to the corresponding same equivalent load. Since the volta ge outputs were almost same at steady

PAGE 54

41 stage when connected to the same equivalent load, with 10000 Hz Figure 4.25: MFC Energy Chart under Same Load Duty Ratio. switching frequency the reactor can be extracted for longer time, so the energy outputs with 10000 Hz switching frequency were higher. In conclusion, applying higher switching frequency to an MFC load, could result in more energy in most cases.

PAGE 55

42 4 .2.4 Biochemical Values Output during MFC Operation Figure 4.26: MFC pH under Same Load Duty Ratio. Figure 4.27: MFC EC under Same Load Duty Ratio.

PAGE 56

43 Figure 4.28: MFC DO under Same Load Duty Ratio. Figure 4.29: MFC ORP under Same Load Duty Ratio. All four biochemical parameters showed the shame shape as the results from the previous batches, and for pH, DO, ORP values, there was no clear difference found be tween 10000 Hz or 2000 Hz switching frequency. With 10000 Hz switching frequency, higher

PAGE 57

44 s olution conductivity value were showed in the results, which confirmed the result in the pH, DO, ORP values in the MFC reactor, and higher switching frequency may indicate higher solution conductivity. 4.3 MFC under Different Operating Points In the experiments, the MFC reactor was connected to a circuit board, where three different resistors (500 , 375 , 250 ), three different duty ratio (0.75, 0.5, 0.25), and all operated under 10000 Hz switching frequency. The resistance connected to the reactor and duty ratio, switching frequency used in the experiments and the equivalent resistance values are listed in the table below. Table 4.3: Different Load Condition under Different Operating Points. Duty Ratio Resistance Connected( ) Equivalent Resistance( ) 0.25 250 1000 0.25 375 1500 0.25 500 2000 0.50 250 500 0.50 375 750 0.50 500 1000 0.75 250 333 0.75 375 500 0.75 500 666

PAGE 58

45 4.3.1 MFC Voltage and Power Output In order to evaluate the performance of the reactor under different operating points and different load conditions, a polarization test was examined as control group. Polarization curves were obtained by the single cycle method [36], various external resistances were connected across the MFC, with each resistance being connected for several minutes and the voltage recorded using a circuit board and computer. The polarization curves is in Figure 4.3 0 an d Figure4.31 . Figure 4.30: MFC Polarization Curve.

PAGE 59

46 Figure 4.31 : MFC Polarization Curve Part. In order to evaluate the actual experimental data, the median value of voltage and power were calculated from the stable stage of each batch of the experiment which can represent the reactor performance and the value were listed in the Table4.4 below. Table 4.4: Duty ratio, switching frequency and equivalent resistance values. V m ( mv ) I ( mA ) P ( µw ) R eq ( ) 435 0.2175 95 2000 406 0.2707 110 1500 352 0.3520 123 1000 340 0.4533 154 750 324 0.4865 157 666 307 0.6140 189 500 262 0.7868 206 333 The MFC power is calculated by equation3.4, a nd MFC current is calculated by the equation,

PAGE 60

47 , (4.2) Where Vm is the median value and R eq is the equivalent resistance value from the Table4.4. Figure 4.32: MFC Test Data Curve. Comparing the Figure 4.31, and Figure 4.32, drafted by data from polarization test and data from the experiment batches, the curve drafted by experimental data showed the same trend as in part of the polarization test, but both voltage and power values were lower compared to the results in polarization test. This suggested that when the resistance were applied with duty ratio and switching frequency, the results from different operating points follow the polarization test, but lower voltage output and lower power output were likely to be produced compared to the equivalent fixed loads.

PAGE 61

48 4.3.2 MFC Biochemical Parameters The median value of four biochemical paramete r s were calculated from the sta ble stage of each batch of the experiment which can represent the actual reactor performance and the value were listed in the table below. Table 4.5: Biochemical Values under Different Load Conditions. R eq ( ) pH EC( s/cm ) DO( mg/l ) ORP Absolute value( mv ) 2000 8.217 9143 0.81 493 1500 8.320 10210 1.07 517 1000 8.870 11690 0.84 538 750 8.350 10010 0.94 499 666 8.046 8878 0.75 489 500 7.844 10470 0.45 512 333 7.934 9452 0.14 425 From Figure 4.33 to Figure 4.33, all biochemical parameters were drafted against the equivalent loads .

PAGE 62

49 Figure 4.33: Average MFC pH under Different Equivalent Load. Figure 4.34: Average MFC EC under Different Equivalent Load.

PAGE 63

50 Figure 4.35: Average MFC DO under Different Equivalent Load Figure 4.36: Average MFC ORP under Different Equivalent Load. T he figure of biochemical parameter showed certain relations between the external loads. With s/cm to 12 000 s/cm , the ORP value range from 540 mv to 480 mv . Only DO value followed the voltage output, the lower the DO values were, the higher MFC voltage were, and also value showed a linear increase when the equivalent loads were under 750 . It indicates that the oxygen in the reactor affects the voltage production and DO value may change with the

PAGE 64

51 external load when the resistance is relatively low. So we can conclude that there were no linear relations between external load and the value of solution conductivity ORP, the values were consistent between certain ranges. The solution pH value changes but not related to the external load, with 1000 equivalent load, reached the highest value of 8.9. The dissolved oxygen the DO value was inversel y proportional to the MFC output voltage.

PAGE 65

52 CHAPTER V CONCLUSION In conclusion, under the same initial condition when the MFC reactor is connected to the fixed load, the bacteria can produce voltage for a longer time, compared to when connected to the load applied with switching frequency and duty ratio. Also when MFC reactor is connected to the fixed load, it can produce more overall energy. The switching frequency and duty ratio dont have any linear relations with pH, DO, ORP value. These bio chemical var iables mostly depend on the reactors solution condition. The higher switching frequency results in higher EC value and longer MFC electricity producing time in most cases. The pH, EC, ORP values dont affect by changes of equivalent external load values. Th e DO value has an inversely proportional relations with of equivalent external load values when the load is low.

PAGE 66

53 REFERENCES [1]Richard S Berk and James H Canfield. Bioelectrochemical energy conversion. Applied microbiology, 12(1):10 12, 1964. [2]John B Davis and Harold F Yarbrough. Preliminary experiments on a microbial fuel cell. Science, 137(3530):615 616, 1962. [3]Korneel Rabaey, Nico Boon, Steven D Siciliano, Marc Ve rhaege, and Willy Ver straete. Biofuel cells select for microbial consortia that se lf mediate electron transfer. Applied and environmental microbiology, 70(9):5373 5382, 2004. [4]Daniel R Bond and Derek R Lovley. Electricity production by geobacter sul furreducens attached to electrodes. Applied and environmental microbiology, 69(3):154 8 1555, 2003. [5]Yuri A Gorby, Svetlana Yanina, Jeffrey S McLean, Kevin M Rosso, Dianne Moyles, Alice Dohnalkova, Terry J Beveridge, In Seop Chang, Byung Hong Kim, Kyung Shik Kim, et al. Electrically conductive bacterial nanowires produced by shewanella on eidensis strain mr 1 and other microorganisms. Proceedings of the National Academy of Sciences, 10 3(30):11358 11363, 2006. [6]Bruce E Logan. Peer reviewed: extracting hydrogen and electricity from renew able resources, 2004. [7]Korneel Rabaey, Nico Boon, Monica H¨ofte, and Willy Verstraete. Microbial phenazine production enhances electron transfer in biofuel cells. Environmental science & technology, 39(9):3401 3408, 2005.

PAGE 67

54 [8]Shaoan Cheng and Bruce E Logan. Sustainable and efficient biohydrogen produc tion via electrohydrogenesis. Proceedings of the National Academy of Sciences, 104(47):18871 18873, 2007. [9]Bruce E Logan. Microbial fuel cells. John Wiley & Sons, 2008. [10]Bruce E Logan, Bert Hamelers, Ren´e Rozendal, Uwe Schro¨der, Ju¨rg Keller, Ste fano Freguia, Peter Aelterman, Willy Verstraete, and Korneel Rabaey. Microbial fuel cells: methodology and technology. Environmental science & technology, 40(17):5181 5192, 2006. [11]Hong Liu and Bruce E Logan. Electricity generation using an air cathode single chamber microbial fuel cell in the presence and absence of a proton exchange membrane. Environmental science & technology, 38(14):4040 4046, 2004. [12]Yanzhen Fan, Hongqiang Hu, and Hong Liu. Enhanced coulombic efficiency and power densi ty of air cathode microbial fuel cells with an improved cell configu ration. Journal of Power Sources, 171(2):348 354, 2007. [13]Eric A Zielke. Design of a single chamber microbial fuel cell. Retrieved from Humboldt State University, School of Engineering Web site: http://www. engr. psu. edu/ce/enve/logan/bioenergy/pdf/Engr 499 final zielke. pdf, 2005. [14]Daniel R Bond, Dawn E Holmes, Leonard M Tender, and Derek R Lovley. Electrode reducing microorganisms that harvest energy from marine sediments. Scienc e, 295(5554):483 485, 2002.

PAGE 68

55 [15]Korneel Rabaey, Geert Lissens, Steven D Siciliano, and Willy Verstraete. A microbial fuel cell capable of converting glucose to electricity at high rate and effi ciency. Biotechnology letters, 25(18):1531 1535, 2003. [16]Boo ki Min, Shaoan Cheng, and Bruce E Logan. Electricity generation using membrane and salt bridge microbial fuel cells. Water research, 39(9):1675 1686, 2005. [17]SangEun Oh, Booki Min, and Bruce E Logan. Cathode performance as a factor in electricity generat ion in microbial fuel cells. Environmental science & technology, 38(18):4900 4904, 2004. [18]Shaoan Cheng, Hong Liu, and Bruce E Logan. Increased power generation in a continuous flow mfc with advective flow through the porous anode and reduced electrode s pacing. Environmental science & technology, 40(7):2426 2432, 2006. [19]Benyi Xiao, Fang Yang, and Junxin Liu. Enhancing simultaneous electricity production and reduction of sewage sludge in two chamber mfc by aerobic sludge digestion and sludge pretreatme nts. Journal of hazardous materials, 189(1):444 449, 2011. [20]Qian Deng, Xinyang Li, Jiane Zuo, Alison Ling, and Bruce E Logan. Power generation using an activated carbon fiber felt cathode in an upflow microbial fuel cell. Journal of Power Sources, 195( 4):1130 1135, 2010. [21]Ananta Lakshmi Kothapalli. Sediment microbial fuel cell as sustainable power resource. PhD thesis, The University of Wisconsin Milwaukee, 2013. [22]Peter Aelterman, Korneel Rabaey, Hai The Pham, Nico Boon, and Willy Ver straete.

PAGE 69

56 C ontinuous electricity generation at high voltages and currents using stacked microbial fuel cells. Environmental science & technology, 40(10):3388 3394, 2006. [23]Minghua Zhou, Meiling Chi, Jianmei Luo, Huanhuan He, and Tao Jin. An overview of electrode materials in microbial fuel cells. Journal of Po wer Sources, 196(10):4427 4435, 2011. [24]Bruce Logan, Shaoan Cheng, Valerie Watson, and Garett Estadt. Graphite fiber brush anodes for increased power production in air cathode microbial fuel cells. Enviro nmental science & technology, 41(9):3341 3346, 2007. [25]Sang Eun Oh and Bruce E Logan. Proton exchange membrane and electrode sur face areas as factors that affect power generation in microbial fuel cells. Applied microbiology and biotechnology, 70(2):16 2 169, 2006. [26]Cristian Picioreanu, Ian M Head, Krishna P Katuri, Mark CM van Loosdrecht, and Keith Scott. A computational model for biofilm based microb ial fuel cells. Water Research, 41(13):2921 2940, 2007. [27]Cristian Picioreanu, KP Katuri, IM Head, Mark CM van Loosdrecht, and Keith Scott. Mathematical model for microbial fuel cells with anodic biofilms and anaerobic digestion. Water science and technology, 57(7):965 971, 2008. [28]Krishna P Katuri, Keith Scott, Ian M Head, Cristian Picioreanu, and To m P Curtis. Microbial fuel cells meet with external resistance. Bioresource technology, 102(3): 2758 2766, 2011.

PAGE 70

57 [29]Barry P Rosen and Simon Silver. Ion transport in prokaryotes. Academic Press, 2014. [30]Eric R Olson. Influence of ph on bacterial gene expr ession. Molecular microbi ology, 8(1):5 14, 1993. [31]S Venkata Mohan, G Mohanakrishna, B Purushotham Reddy, R Saravanan, and PN Sarma. Bioelectricity generation from chemical wastewater treatment in mediatorless (anode) microbial fuel cell (mfc) using se lectively enriched hydro gen producing mixed culture under acidophilic microenvironment. Bioc hemical Engineering Journal, 39(1):121 130, 2008. [32]Geun Cheol Gil, In Seop Chang, Byung Hong Kim, Mia Kim, Jae Kyung Jang, Hyung Soo Park, and Hyung Joo Kim. O perational parameters affecting the per formannce of a mediator less microbial fuel cell. Biosensors and Bioelectronics, 18(4):327 334, 2003. [33]Zhuwei Du, Haoran Li, and Tingyue Gu. A state of the art review on micro bial fuel cells: a promising techno logy for wastewater treatme nt and bioenergy. Biotechnology advances, 25(5):464 482, 2007. [34]Joo Youn Nam, Hyun Woo Kim, Kyeong Ho Lim, Hang Sik Shin, and Bruce E Logan. Variation of power generation at different buffer types and conductivities in single chamber microbial fuel cells. Biosensors and Bioelectronics, 25(5):1155 1159, 2010. [35]SE Oh, JR Kim, J H Joo, and BE Logan. Effects of applied voltages and dis solved oxygen on sustained power generation by microbial fuel cells. Water science and tech nology, 60(5):1311 1317, 2009.

PAGE 71

58 [36]Valerie J Watson and Bruce E Logan. Analysis of polarization methods for elimination of power overshoot in microbial fuel cells. Electrochemistry Commu nications, 13(1):54 56, 2011.