Embed Size (px)
Citation preview
The Pennsylvania State University
The Graduate School
Department of Civil and Environmental Engineering
NUTRIENT AND HEAT RECOVERY FROM WASTE STREAMS USING
MICROBIAL ELECTROCHEMICAL TECHNOLOGIES
A Dissertation in
Environmental Engineering
by
Roland D. Cusick
© 2013 Roland D. Cusick
Submitted in Partial Fulfillment of the Requirements
for the Degree of
Doctor of Philosophy
December 2013
The dissertation of Roland D. Cusick was reviewed and approved* by the following:
Bruce E. Logan Evan Pugh Professor of Engineering, Kappe Professor of Environmental Engineering Dissertation Advisor Chair of Committee
John M. Regan Professor of Environmental Engineering
Brian A. Dempsey Professor of Environmental Engineering
Michael A. Hickner Associate Professor of Material Science and Engineering
Peggy Johnson Professor of Civil Engineering Head of the Department of Civil and Environmental Engineering
*Signatures are on file in the Graduate School
iii
ABSTRACT
Microbial Electrochemical Technologies (METs) are an emerging design platform
capable of converting organic matter directly into electrical energy. Two new METs were
developed to enhance the sustainability of wastewater treatment. The first set of systems
utilizes electrical and ionic currents to couple bio-electricity generation from wastewater
organics with the recovery of wastewater nutrients via salt crystallization. The second set
of METs was designed to enable the recovery of waste heat using engineered salinity
gradients.
The first system was an energy efficient method of recovering both energy and
nutrients from wastewater in a microbial electrolysis cell (MEC) by producing hydrogen
to increase solution pH and precipitate struvite (MgNH4PO4·6H2O). The concept was first
proven in a single-chamber MEC in which hydrogen evolution cathodes were either
stainless steel 304 mesh or flat plates. The cathodes accumulated a layer of crystals,
which were verified as struvite using a scanning electron microscope capable of energy
dispersive spectroscopy (SEM-EDS). Phosphate removals were low (20 – 38%) because
the pH increase was localized to the cathode surface. Energy produced as hydrogen was
higher than the energy invested to drive hydrogen evolution, implying that MECs may be
useful both as a method for hydrogen gas production and struvite production.
In order to increase phosphate removal and prevent cathode scale accumulation, a
two-chamber MEC was designed where the cathode operated as a fluidized bed
crystallization reactor. Within the cathode, hydrogen production continuously raised
cathode solution pH, and a fluidized bed of seed particles provided surface area for
iv
struvite crystal growth while scoring the cathode surface. When MEC electrodes
generated current, the cathode solution pH was maintained between 8.3 – 8.8 under
continuous flow conditions, and soluble phosphorus removal ranged from 70 – 85%, as
compared to 10 – 20% when the fluidized bed was operated under open circuit
conditions. At low current densities (≤ 2 mA/m2), scouring by fluidized particles
prevented cathodic scale formation for the duration of continuous flow experiments. At
an applied voltage of 1.0 V, energy consumption from the power supply and pumping
was 0.2 Wh/L (7.5 Wh/g-P). This energy requirement could be fully offset if future
hydrogen recoveries could exceed 80%. These results indicate that a fluidized bed
cathode MEC is a promising method of sustainable electrochemical nutrient recovery.
The second type of MET was developed for converting waste heat and wastewater
organics into electricity using ammonium bicarbonate (AmB) salt solutions in a microbial
reverse electrodialysis cell (MRC). A MRC couples microbial fuel cell (MFC) electrodes
with a reverse electrodialysis (RED) membrane stack, a technology capable of recovering
energy released when salt and fresh water mix. By feeding the RED stack salt solutions
composed of AmB, which decomposes to ammonia and carbon dioxide gas at low
temperatures (40 – 60°C), MRCs can create electricity from abundant sources of low-
grade thermal energy such as waste heat. The maximum power density using acetate
reached 5.6 ± 0.04 W/m2-cathode surface area, which was five times that produced
without the dialysis stack, and 3.0 ± 0.05 W/m2 with domestic wastewater. The greatest
contribution to MRC power came from enhanced MFC electrode performance, which
increased by more than 300% with acetate and over 700% with domestic wastewater.
v
MFC electrode power production was further investigated in MRCs having a minimal
number of membranes. It was observed that using only a single RED cell pair (CP),
created by operating the cathode in concentrated AmB, dramatically increased power
production normalized to cathode area from both acetate and wastewater. Galvanostatic
electrochemical impedance spectroscopy indicated that power increases in the 1-CP MRC
were caused by reductions in solution and kinetic resistances at the cathode. The addition
of a second RED cell pair further increased power production and anode biofilm activity.
Power densities achieved here were close to those previously achieved using 11
membranes, indicating near optimal electrode performance with only one or two cell
pairs. When normalized to total membrane area, the minimal cell pair MRC power
densities was 4 – 10 times higher than previous MRCs. The rate of wastewater COD
removal, normalized to reactor volume, was 30 – 50 times higher in 1-CP and 2-CP
MRCs than that in a single-chamber MFC. These findings showed that even a single cell
pair AmB RED stack could significantly enhance electrical power production and
wastewater treatment rates.
vi
TABLE OF CONTENTS
List of Figures .............................................................................................................. ix
List of Tables ............................................................................................................... xiii
Acknowledgements ...................................................................................................... xiv
Chapter 1 General Introduction ................................................................................... 1
1.1 Dissertation Scope and Outline ................................................................................... 2 1.2 Additional Research Publications ............................................................................... 6 1.3 Literature Cited ........................................................................................................... 8
Chapter 2 Literature Review ........................................................................................ 10
2.1 Microbial Electrochemical Technologies ................................................................... 10 2.1.1 MFC and MEC Basic Principles ...................................................................... 10 2.1.2 Anode Materials ............................................................................................... 12 2.1.3 Cathode Materials ............................................................................................. 13 2.1.4 Membranes ....................................................................................................... 13 2.1.5 Internal Resistance ........................................................................................... 14 2.2 Phosphorus Removal and Recovery from Wastewater ............................................... 15 2.2.1 Introduction ...................................................................................................... 15 2.2.2 Phosphorus Management at Wastewater Treatment Plants ............................. 17 2.2.3 Phosphorus Precipitation from Nutrient Rich Side-streams ............................. 18 2.2.4 Electrochemical Phosphate Salt Precipitation .................................................. 23 2.3 Reverse Electrodialysis ............................................................................................... 24 2.3.1 Introduction ...................................................................................................... 24 2.3.2 Reverse Electrodialysis Power Production ....................................................... 27 2.3.3 Membrane Resistance ....................................................................................... 28 2.3.4 Solution Resistance .......................................................................................... 29 2.3.5 Spacers .............................................................................................................. 30 2.3.6 Redox Electrode Pairs ...................................................................................... 30 2.4 Literature Cited ........................................................................................................... 33
Chapter 3 Energy Efficient Phosphate Recovery as Struvite Within a Single Chamber Microbial Electrolysis Cell ................................................................... 44
3.1 Abstract ....................................................................................................................... 44 3.2 Introduction ................................................................................................................. 45 3.3 Materials and Methods ................................................................................................ 47 3.3.1 Reactor Construction and Operation ................................................................ 47 3.3.2 Analytical Techniques ...................................................................................... 49 3.4 Results and Discussion ............................................................................................... 50 3.4.1 Cathode Crystal Accumulation and Phosphate Removal ................................. 50
vii
3.4.2 Crystal Analysis ............................................................................................... 51 3.4.3 MEC Performance ............................................................................................ 52 3.4.4 Electrochemical Impedance Analysis of Cathodes .......................................... 55 3.5 Conclusions ................................................................................................................. 56 3.6 Figures ........................................................................................................................ 57 3.7 Literature Cited ........................................................................................................... 62
Chapter 4 Electrochemical Struvite Precipitation from Digester Effluent with a Fluidized Bed Cathode Microbial Electrolysis Cell ............................................. 65
4.1 Abstract ....................................................................................................................... 65 4.2 Introduction ................................................................................................................. 66 4.3 Materials and Methods ................................................................................................ 69 4.3.1 Reactor Construction ........................................................................................ 69 4.3.2 Solutions ........................................................................................................... 70 4.3.3 Reactor Operation ............................................................................................. 70 4.3.4 Solution and Electrochemical Measurements .................................................. 71 4.3.5 Analysis ............................................................................................................ 72 4.4 Results and Discussion ............................................................................................... 73 4.4.1 Phosphorus Removal in the Fluidized Bed Cathode ........................................ 73 4.4.2 Phosphorus Precipitation in the Fluidized Bed Cathode .................................. 74 4.4.3 Cathode Scaling ................................................................................................ 75 4.4.4 Molar Ionic Removal in the Fluidized bed Cathode ........................................ 76 4.4.5 Electrolyte Modeling ........................................................................................ 77 4.4.6 COD Removal in the Anode Chamber ............................................................. 78 4.4.7 Energy Consumption ........................................................................................ 78 4.5 Conclusions ................................................................................................................. 79 4.6 Tables .......................................................................................................................... 81 4.7 Figures ........................................................................................................................ 83 4.8 Literature Cited ........................................................................................................... 92
Chapter 5 Energy Capture from Thermolytic Salt Solutions in Microbial Reverse Electrodialysis Cells ............................................................................................. 97
5.1 Abstract ....................................................................................................................... 97 5.2 Introduction ................................................................................................................. 98 5.3 Materials and Methods ................................................................................................ 100 5.3.1 Reactor Construction ........................................................................................ 100 5.3.2 Solutions ........................................................................................................... 101 5.3.3 Power Measurement ......................................................................................... 102 5.3.4 Analysis ............................................................................................................ 103 5.4 Results and Discussion ............................................................................................... 104 5.4.1 Effect of AmB Concentration on MRC Power Production .............................. 104 5.4.2 Effect of Salinity Ratio and Flow Rate on MRC Power Production ................ 106 5.4.3 Energy Production in Batch Recycle Experiments .......................................... 106 5.4.4 Energy Recovery and Efficiency ...................................................................... 107 5.4.5 Ammonia Transport into the Anode ................................................................. 108 5.4.6 Power Production from Domestic Wastewater ................................................ 108 5.5 Conclusions ................................................................................................................. 109
viii
5.6 Figures ........................................................................................................................ 110 5.7 Literature Cited ........................................................................................................... 118
Chapter 6 Minimal RED Cell Pairs Markedly Improve Electrode Kinetics and Power Production in Microbial Reverse Electrodialysis Cells ............................. 122
6.1 Abstract ....................................................................................................................... 122 6.2 Introduction ................................................................................................................. 123 6.3 Materials and Methods ................................................................................................ 125 6.3.1 Reactor Construction ........................................................................................ 125 6.3.2 Solutions ........................................................................................................... 127 6.3.3 Electrochemical Measurements and Analysis .................................................. 128 6.4 Results and Discussion ............................................................................................... 130 6.4.1 MRC Power Production ................................................................................... 130 6.4.2 Electrode and RED Stack Potential Losses in MRCs ...................................... 131 6.4.3 Anolyte Composition and Membrane Transport Resistance ............................ 133 6.4.4 Effect of RED Stack Potential on Bio-Anode Charge Transfer Resistance ..... 135 6.4.5 Wastewater Treatment Rates and Energy Recovery ........................................ 135 6.5 Outlook ....................................................................................................................... 137 6.6 Tables .......................................................................................................................... 138 6.7 Figures ........................................................................................................................ 139 6.8 Literature Cited ........................................................................................................... 143 Appendix A Energy Efficient Phosphate Recovery as Struvite Within a Single
Chamber Microbial Electrolysis Cell ....................................................................... 146 Appendix B Energy Production from Thermolytic Solutions in Microbial Reverse
Electrodialysis Cells ................................................................................................. 149 Appendix C Minimal RED Cell Pair Microbial Reverse Electrodialysis Cells ............... 152
ix
LIST OF FIGURES
Figure 2-1: Diagram of microbial electrochemical technologies. .......................................... 12
Figure 2.2: Examples of scale accumulation in pipes and crystallized phosphate salts recovered from a fluidized bed reactor. ........................................................................... 21
Figure 2.3: Diagram of a commercial scale fluidized bed reactor for struvite precipitation. .................................................................................................................... 22
Figure 2.4: Diagram of a reverse electrodialysis reactor. ....................................................... 26
Figure 3.1: Single chamber MESC reactor schematic. ........................................................... 57
Figure 3.2: Struvite Crystal Growth and phosphate removal in MESC with stainless steel mesh (SSM) and shim (SSF) cathodes. ............................................................................ 57
Figure 3.4: (a) Hydrogen production rate (Q) and (b) current density of single chamber MEC (C) and MESC (S) reactors. ................................................................................... 59
Figure 3.5: (a) Coulombic Efficiency (CE), (b) electrical efficiency (ηe) and (c) overall efficiency (ηe+s) of single chamber MEC (C) and MESC (S) reactors. ........................... 60
Figure 3.6: Energy comsumption and recovery for SSM and SSF in MEC (C) and MESC (S) reactors a) 0.75 b) 0.90 and c) 1.05 V. ....................................................................... 61
Figure 4.1: Process flow of fluidized bed cathode microbial electrolysis cell with close up of electrode interface. .................................................................................................. 83
Figure 4.2: a) Molar ionic removal and b) soluble (sP) and total (tP) phosphorus removal in the fluidized bed cathode. ............................................................................................ 84
Figure 4.3: a) MEC current density (b) cathode pH and (c) effluent soluble P concentrations during eight day continuous flow experiments. ....................................... 85
Figure 4.4: a) Phosphorus precipitation and collection rate in the fluidized bed cathode and (b) energy input normalized to precipitated and collected phosphorus. .................... 86
Figure 4.5: SEM images of stainless steel mesh cathodes operated at a) open circuit, b) 0.8 V, c) 1.0 V and d) 1.4 V. ............................................................................................ 87
Figure 4.6: a) Ionic, (b) ammonium and (c) COD removal in the anode chamber. ................ 88
Figure 4.7: Modeled energy balance and a function of cathodic hydrogen recovery. ............ 89
Figure 4.8: Modeled energy input predictions as a function of a) applied voltage and b) influent phosphorus concentration. .................................................................................. 90
Figure 4.9: Electrolyte modeling of a) equivalent NaOH addition and b) super-saturation of sparingly soluble salts, and c) ionization in the digestate. ........................................... 91
x
Figure 5.1: (a) Main components of the microbial reverse electrodialysis cell (MRC), showing the membrane stack between the electrodes, the reference electrodes and the circuit containing a load (resistor). (b) Example of how the anion- (AEM) and cation- (CEM) exchange membranes are used to selectively drive the flow of positive ions to the right (towards the cathode) and the negatively-charged ions to the left (towards the anode). (c) Expanded view of the membrane stack showing flow path of the high (HC) and low (LC) concentrate solutions of ammonium bicarbonate. (d) Construction of the gaskets used to direct the flow from one LC chamber to the next LC chamber, avoiding the HC chamber through a short flow path through the membrane and gasket. ........................................................................... 110
Figure 5.2: Peak power density of MRC as well as the contributions to total power from the RED stack, and MFC electrodes at various HC concentrations. The dashed line represents peak power density of the same electrodes in a single chamber MFC. .......... 111
Figure 5.3: Polarization curves of the MRC at various HC concentrations as well and the polarization curves of the electrodes in a single chamber MFC. ..................................... 112
Figure 5.4: a) RED stack potential as well as b) anode (filled symbols) and cathode (open symbols) potentials, measured during polarization curves, in the single chamber MFC (SC) and the three highest HC concentrations. ........................................ 113
Figure 5.5: Salinity ratio vs. peak power density at HC concentration of 0.95 M and flow rate of 1.6 mL/min. Dashed line represented peak power density of single chamber MFC fed identical substrate. ............................................................................................ 114
Figure 5.6: a) Energy recovery, efficiency and b) and energy balance of the MRC vs. HC concentration. ................................................................................................................... 115
Figure 5.7: a) Peak power density and (b) anode and cathode potentials of MRC and single chamber MFC fed domestic wastewater. .............................................................. 116
Figure 5.8: Batch recycle component (MFC, RED and total MRC) power profile of MRC fed domestic wastewater. ....................................................................................... 117
Figure 6.1: Schematics of tested reactor configurations. ........................................................ 141
Figure 6.2: (a,b) Power densities normalized to cathode area, (c,d) anode and cathode potentials (vs. SHE), and (e,f) reverse electrodialysis (RED) stack potentials of MFCs, 1-CP and 2-CP MRCs fed (filled markers) bicarbonate buffered acetate and (open markers) domestic wastewater. .............................................................................. 141
Figure 6.3: Internal resistance of 1-CP reactors components fed raw domestic wastewater (WW), buffered wastewater (WW-HC), and buffered acetate (Ac), determined from (a) direct current (DC) polarization and (b) alternating current galvanostatic electrochemical impedance spectroscopy. ....................................................................... 142
Figure 6.4: Internal resistance of MFC, 1-CP, and 2-CP MRC reactors components fed buffered acetate (Ac) determined from (a) direct current (DC) polarization and (b) alternating current (AC) galvanostatic electrochemical impedance spectroscopy. ......... 142
xi
Figure 6.5: The relative effects of reduced internal resistance and RED stack potential on power enhancement in 1-CP and 2-CP MRCs fed wastewater (WW) and acetate (Ac). ................................................................................................................................. 143
Figure 6.6: a) Volumetric COD removal rates and (b) energy recovery from wastewater and acetate in batch recycle experiments. ........................................................................ 144
Figure 6.7: Estimation of 1-CP and 2-CP MRC power density with buffered acetate (Ac) and wastewater (WW) normalized to (a) total membrane area and (b) cathode area versus reduction in RED stack resistance. ................................................................ 145
Figure A.1: Concurrent LSV (filled) and pH (open) measurements at the surface of SSM cathode in 3mM PBS and 75 mM HCO3
- electrolyte. ...................................................... 150
Figure A.2: Average batch time and crystal accumulation of MESC reactors with both SSF and SSM cathodes at applied voltages of 0.75, 0.90 and 1.05 V. ............................ 151
Figure A.4: EIS determination of internal resistance for SSM and SSF in MEC (C) and MESC (S) reactors at cathode potentials observed in fed-batch operation at a) 0.75 b) 0.90 and c) 1.05 V. ....................................................................................................... 152
Figure B.1: Relationship between ammonium bicarbonate concentration and solution. Figure A.3: Anode (an) and cathode (cat) potentials measured in fed-batch operation at a) 0.75 b) 0.90 and c) 1.05 V. ...................................................................... 153
Figure B.2: Relationship between total ammonium bicarbonate salt concentration and species concentration at pH =7 and T = 25 °C as estimated with OLI stream analyzer software. ........................................................................................................................... 154
Figure B.3: Relationship between species concentration and species activity at pH =7 and T = 25 °C as estimated with OLI stream analyzer software. ..................................... 154
Figure B.4: Power density curves of the MRC (HC = 0.95 M, SR = 100) at different salt solution flow rates. ........................................................................................................... 155
Figure B.5: Batch recycle component (MFC, RED and total MRC) power profile of MRC fed sodium acetate and operating with an external resistance of 300 Ω. ............... 155
Figure C.1: Internal resistance determined from the slope of linear polarization curves for the MFC, 1-CP and 2-CP MRC fed acetate (Ac) and domestic wastewater (WW). .. 156
Figure C.2: a) Power density, (b) electrode potentials, and (c) stack potentials of 1-CP reactor fed acetate (Ac), high conductivity wastewater (WW-HC) and raw wastewater (WW). ........................................................................................................... 157
Figure C.3: Nyquist plots of GEIS data collected from 1-CP MRCs fed acetate, wastewater, and high conductivity wastewater. Impedance spectra were simultaneously collected for (a) whole cell and (b) anode (c) RED stack and (d) cathode. ............................................................................................................................ 157
xii
Figure C.4: Nyquist plots of GEIS data collected from electrode and RED of MFC and MRCs fed acetate. Impedance spectra were simultaneously collected for (a) anode (c) RED stack and (d) cathode. ........................................................................................ 158
Figure C.5: Nyquist plots with equivalent circuit models and fits of GEIS data collected from the (a) anode (c) RED stack and (d) cathode of 1-CP MRC fed acetate. ................ 158
xiii
LIST OF TABLES
Table 2.1: Molecular formulas and formation constants for phosphate and carbonate salts. .................................................................................................................................. 19
Table 4.1: Common name, formula and solubility constant of sparing soluble salts with potential to form in digester effluent. ............................................................................... 81
Table 4.2: Solution characteristics of digestate fed to cathode and anode chambers. ............ 81
Table 4.3: Cathode effluent ionic concentrations. .................................................................. 82
Table 4.4: Molar composition of collected solids as determined by acid dissolution. ........... 82
Table 6.1: Anode substrate characteristics .............................................................................. 138
xiv
ACKNOWLEDGEMENTS
I would like to thank Dr. Bruce E. Logan, for providing me this opportunity to pursue
a graduate degree at Penn State as well as a myriad of life altering experiences over the
last five years. His encouragement and constructive feedback have pushed me to become
a better researcher, communicator and engineer. I would also like to thank Dr. John M.
Regan, Dr. Michael Hickner and Dr. Brian A. Dempsey for their support and serving on
my committee. I would like to acknowledge my fellow researchers at the H2E center. In
particular I want to thank David Jones, Dr. Younggy Kim, for his research guidance,
Mark Ullery for data collection assistance, as well as Marta Hatzell and Fang Zhang for
their assistance in experimental design and data analysis.
I am eternally grateful to my wife, Elise Belknap, for providing me with the
unconditional love, compassion and support needed to complete my graduate education.
Without the guidance and encouragement of my family, I would have never embarked on
this journey.
Finally, I would like to thank the King Abdullah University of Science and
Technology for financially supporting my research.
Chapter 1
General Introduction
If properly managed, wastewater can sustainably provide society with three of the
most critical resources of the 21st century: energy, nutrients and water. Each day roughly
126 billion liters of wastewater are treated in the United States [1]. Contained within this
out-flow are roughly 240 mega-watt hours (MWh) of energy in the form of organic
matter, 880 MWh of energy as heat, 5 metric tons of nitrogen, and 1 metric ton of
phosphorus [2]. Although wastewater contains nine times the energy required for
treatment [3], current wastewater treatment practices consumes 3% of the national energy
grid [4].
Microbial electrochemical technologies (METs) have attracted much attention
recently due to the unique ability to convert soluble organic matter directly into electrical
energy. Similar to fuel cells and batteries, METs are composed of an anode, where
electrical current is generated, and a cathode, where current is consumed. At the anode,
dissimilatory metal reducing bacteria (commonly referred to as exoelectrogens [5])
oxidize the organic matter present in wastewater and continuously transfer electrons to
the circuit. At MET cathodes, electrical current is consumed to reduce oxygen to water
(MFC) [6], or protons to hydrogen gas (MEC) [7]. In a MFC, electricity generation is
spontaneous because the oxidation of organic matter occurs at a negative potential (− 0.3
V vs. SHE for acetate) and the potential of oxygen reduction to water is positive
2
(thermodynamic potential of 0.8 V vs. SHE, in practice 0.2 V vs. SHE) [8]. In order to
generate hydrogen at an anaerobic cathode in a MEC, an additional input of electrical
energy is needed (0.4 to 1.0 V in practice); however, the required input is significantly
lower than that required for traditional electrolytic hydrogen generation technologies [9].
Although using METs to directly generate electrochemical energy from wastewater is
an exciting prospect, low power densities and high material costs have limited
commercialization efforts. In comparison, anaerobic digestion (AD), an established
biotechnology, converts organic matter to energy dense methane at roughly 1% of the
capital cost of a MFC [10]. However, the ability of METs to drive ionic current creates
the opportunity to leverage wastewater organics towards displacing energy and cost
intensive industrial processes such as pH adjustment [11] and ion separations [12]. It is
shown here that METs can be designed to not only generate electrical current from
soluble organic matter but also recover solid-phase nutrients from wastewater and
transform low-grade thermal energy into electricity.
1.1 Dissertation Scope and Outline
To increase the potential sustainability of wastewater treatment METs, two new
technologies were developed that simultaneously convert organic wastewater into energy
and recover resources such as phosphate, nitrogen and waste heat.
In Chapter 3, I examine nutrient recovery as magnesium ammonium phosphate
hexahydrate (struvite) in a single chamber microbial electrolysis cell. Struvite can be
produced from crystallization of inorganics in nutrient rich animal and domestic
wastewaters during wastewater treatment. Controlled struvite recovery from wastewater
3
can be achieved by raising solution pH via chemical base addition, carbon dioxide
stripping, or electrolysis. In order to make electrochemical struvite precipitation more
energy efficient, I examined production of struvite in a microbial electrolysis cell (MEC).
The objectives of this study were therefore to show that struvite could be precipitated at
the cathode of a MEC, and to quantify the affect applied potential and cathode material
had on system metrics such as struvite precipitation, phosphate removal, hydrogen
production and energy efficiency. The results of this chapter are summarized in a paper
by R. D. Cusick and B. E. Logan entitled “Phosphate recovery as struvite within a single
chamber microbial electrolysis cell” published in the journal Bioresource Technology. I
conducted all experiments and analysis and prepared the first manuscript draft.
In Chapter 4, I addressed two of the deficiencies observed for phosphate recovery in
a single chamber MEC (chapter 3) of low P removal rates and cathode scaling by
constructing a continuous flow, two-chambered MEC in which the cathode
simultaneously operates as a hydrogen evolution chamber and high pH fluidized bed
crystallization chamber. Fluidized bed crystallization reactors (FBRs) are designed to
rapidly recover phosphate salts and reduce loss of colloidal fine particles in solution by
providing abundant surface area for molecular growth and agglomeration. The
hypotheses of this study was that a fluidized bed of seed granules could enhance
phosphate salt removal efficiency and maintain MEC current by scouring crystals from
the cathode surface. The FBR-MEC system effectiveness was determined by comparing
phosphate salt removal rates in open circuit voltage operation (no pH adjustment) and
various applied voltages as well as the sustainability of current generation and energy
demand of phosphorus removal. The results of this chapter will be summarized in a paper
4
by R.D. Cusick, M. Ullery, B. A. Dempsey and B.E. Logan entitled “Electrochemical
struvite precipitation from digestate with a fluidized bed cathode microbial electrolysis
cell”. I designed, constructed and operated the reactor as well as performed data analysis
and manuscript preparation. M. Ullery assisted in data collection. B.A. Dempsey assisted
in electrolyte modeling. All co-authors will assist in the final preparation of the
manuscript.
In Chapter 5, I examined the potential of converting waste heat and wastewater
organics into electricity by using ammonium bicarbonate salt solutions in a microbial
reverse electrodialysis cell (MRC). A MRC couples the spontaneous electrode kinetics of
MFCs with salinity gradient energy of a reverse electrodialysis (RED) stack. Previously,
RED technology had been proposed to harness energy released by the mixing of seawater
and fresh water. With RED saline solutions were composed of ammonium bicarbonate
(AmB), which decomposes to ammonia and carbon dioxide gas at low temperatures (40 –
60°C) [13], MRCs can create electricity from abundant sources of low grade thermal
energy such as waste heat (204 GW) [14], solar (120,000 TW) and geothermal (12 TW)
energy [15]. Therefore, in this study, MRC power production using ammonium
bicarbonate saline solutions was investigated. The effects of saline solution
concentration, salinity ratio (HC/LC), and flow rate on power density were examined
with both synthetic substrate containing acetate and primary clarifier effluent from a local
domestic wastewater treatment plant. The results of this chapter were published in the
journal Science as a paper by R.D Cusick, Y. Kim and B.E. Logan entitled “Energy
capture from thermolytic solutions in microbial reverse-electrodialysis cells.” Y. Kim
assisted in reactor construction and experimental design. I performed all experiments and
5
analysis as well as prepared the first draft of the manuscript. All co-authors contributed to
the preparation of the final manuscript.
In Chapter 6, I investigated the source of dramatic MRC electrode power production
increases (reported in chapter 5) by constructing and operating MRCs with a minimal
number of RED stacks. The objectives of this study were to investigate potential catalytic
benefits AmB on oxygen reduction, and determine to what extent MRC performance
could be improved using only a minimal number of RED cell pairs. To study the effects
of the AmB and cell pairs on MRC performance, power production and internal
resistance were examined using three different reactor configurations: (1) a single
chamber MFC with a bicarbonate buffer used as the single electrolyte in contact with
both electrodes; (2) a one cell pair (1-CP) MRC where the anolyte formed an equivalent
low concentration chamber with the bicarbonate buffer, and the catholyte was the
concentrated solution (1 M AmB); and (3) a two cell pair (2-CP) MRC containing an
additional membrane pair containing high and low concentration AmB solutions. Both
sodium acetate (in 50 mM bicarbonate buffer) and domestic wastewater were used as
fuels to investigate the effect of anolyte composition on electrode and RED stack
performance. To quantify the specific effects of AmB catholyte and RED stack potential
on electrode activity, internal resistance was measured using linear polarization and
galvanostatic electrochemical impedance spectroscopy. The results presented in the
chapter will be summarized in a paper by R.D. Cusick, M.C. Hatzell, F. Zhang, and B.E.
Logan entitled “Minimal RED cell pairs markedly improve electrode kinetics and power
production in microbial reverse electrodialysis cells.” I conducted all experiments and
6
analysis and prepared the initial manuscript draft. M.C. Hatzell and F. Zhang assisted in
experimental design. All co-authors will contribute the to preparation of the manuscript.
1.2 Additional Research Publications
The following list includes published research to which I contributed as a co-investigator
but fall outside the scope of this dissertation:
1. Yates, M.D., R.D. Cusick, I. Ivanov, and B.E. Logan, Exoelectrogenic biofilm as a
template for sustainable formation of a catalytic mesoporous structure. Nano Letters,
2013. Submitted.
2. Hatzell, M.C., R.D. Cusick, and B.E. Logan, Capacitive Mixing Power Production
from Salinity Gradient Energy Enhanced through Exoelectrogen-Generated Ionic
Currents. Energy and Environmental Science, 2013. Submitted.
3. Yates, M.D., R. Cusick, and B.E. Logan, Extracellular Palladium Nanoparticle
Production using Geobacter sulfurreducens. ACS Sustainable Chemistry &
Engineering, 2013. doi: 10.1021/sc4000785
4. Zhu, X., Hatzell, M. C., R.D. Cusick, B.E. Logan, Microbial reverse-electrodialysis
chemical-production cell for acid and alkali production. Electrochemistry
Communications, 2013. 31: p. 52-55.
5. Tenca, A., R.D. Cusick, A. Schievano, R. Oberti, and B.E. Logan, valuation of low
cost cathode materials for treatment of industrial and food processing wastewater
using microbial electrolysis cells. International Journal of Hydrogen Energy, 2013.
38(4): p. 1859-1865.
6. Nam, J.-Y., R.D. Cusick, Y. Kim, and B.E. Logan, Nam, J.-Y., et al., Hydrogen
7
generation in microbial reverse-electrodialysis electrolysis cells using a heat-
regenerated salt solution. Environmental Science and Technology-Columbus, 2012.
46(9): p. 5240.
8
1.3 Literature Cited
1. WIN, Water infrastructure now: recommendations for clean and safe water in the 21st
century, 2001.
2. McCarty, P.L., J. Bae, and J. Kim, Domestic Wastewater Treatment as a Net Energy
Producer–Can This be Achieved? Environmental Science & Technology, 2011. 45(17):
p. 7100-7106.
3. Shizas, I. and D.M. Bagley, Experimental determination of energy content of unknown
organics in municipal wastewater streams. J Energy Eng, 2004. 130(2): p. 45-53.
4. EPA, Wastewater Management Fact Sheet, Energy Conservation, 2006. p. 7.
5. Logan, B.E., Exoelectrogenic bacteria that power microbial fuel cells. Nature Reviews
Microbiology, 2009. 7(5): p. 375-381.
6. Liu, H., R. Ramnarayanan, and B.E. Logan, Production of electricity during wastewater
treatment using a single chamber microbial fuel cell. Environmental Science &
Technology, 2004. 38(7): p. 2281-2285.
7. Liu, H., S. Grot, and B.E. Logan, Electrochemically assisted microbial production of
hydrogen from acetate. Environ Sci Technol, 2005. 39(11): p. 4317-4320.
8. Logan, B.E., Microbial fuel cells2008, Hoboken, NJ: John Wiley & Sons, Inc. 300.
9. Logan, B.E., et al., Microbial electrolysis cells for high yield hydrogen gas production
from organic matter. Environmental Science & Technology, 2008. 42(23): p. 8630-8640.
10. Rozendal, R.A., et al., Towards practical implementation of bioelectrochemical
wastewater treatment. Trends in Biotechnology, 2008. 26(8): p. 450-459.
11. Rabaey, K., et al., High Current Generation Coupled to Caustic Production Using a
Lamellar Bioelectrochemical System. Environmental Science & Technology, 2010.
44(11): p. 4315-4321.
9
12. Mehanna, M., et al., Using microbial desalination cells to reduce water salinity prior to
reverse osmosis. Energy & Environmental Science, 2010. 3(8): p. 1114-1120.
13. McGinnis, R.L., J.R. McCutcheon, and M. Elimelech, A novel ammonia-carbon dioxide
osmotic heat engine for power generation. Journal of Membrane Science, 2007. 305: p.
13-19.
14. EIA, U.S., Annual Energy Review 2010, DOE, Editor 2010.
15. Kamat, P.V., Meeting the clean energy demand: Nanostructure architectures for solar
energy conversion. The Journal of Physical Chemistry C, 2007. 111(7): p. 2834-2860.
10
Chapter 2
Literature Review
2.1 Microbial Electrochemical Technologies
Electrical energy production in microbial electrochemical technologies (METs)
occurs in a similar fashion to other more established electrochemical cells, such as fuel
cell and batteries, by coupling an oxidation reaction at an anode terminal with a reduction
reaction at a cathode terminal. METs are unique in that microbes growing on the anode
catalyze the oxidation of soluble organic matter in wastewater to generate current [1].
Several METs have been designed by tailoring the cathode (both abiotic and biotic) to
accomplish a specific reaction such as oxygen reduction [2], proton reduction to produce
hydrogen [3], metal reduction [4, 5], water reduction to produce hydroxide [6], pollutant
reduction [7] and carbon dioxide reduction to produce organic chemicals [8, 9]. The
METs described in this dissertation utilize either oxygen reduction to produce electricity
or proton reduction to produce hydrogen so this section of the literature review will focus
on substrate and material characteristics that affect power and hydrogen production.
2.1.1 MFC and MEC Basic Principles
Both MFCs and MECs rely on anaerobic microbial oxidation of organic matter at the
anode. Acetate oxidation at pH 7, standard temperature and pressure, is shown as an
example:
CH3CHOO- + H2O → HCO3- + CO2 + 8e- + 8H+ Eo = −0.32 V (2.1)
11
Although many microbes have proved capable of generating current at the anode a
MET, the most ubiquitous bacterial strain in mixed culture biofilms fed wastewater
appears to be Geobacter sulfurreducens [10-12]. G. sulfurreducens can generate high
current densities (~10 A/m2) by directly transmitting electrons to the anode surface
through conductive pili [13]. Electrons travel from the anode through an external circuit
and are consumed at the cathode. At the cathode of MFCs, oxygen is reduced to water.
2O2 + 8e- + 8H+ → 4H2O Eo = +0.82 V (2.2)
Electricity production in MFCs is spontaneous because the cathode potential is more
positive than the anode potential, which creates a driving force for current production.
The maximum thermodynamic potential for a MFC fed acetate should be 1.1 V, but due
to significant potential losses at the electrodes. MFC open circuit potentials are typically
between 0.5 – 0.7 V [14]. At the anode, open circuit potential losses (~0.1 V) result from
bacterial energy recovery for reproduction and cell maintenance [15]. The most
significant potential losses (~0.6 V) occur at the cathode and are caused by catalyst
properties, solution pH and conductivity [16].
In MECs, electrons are consumed in the hydrogen evolution reaction at the cathode:
8e- + 8H+ → 4H2 Eo = −0.414 V (2.3)
Since the HER potential is less than the maximum anode potential, supplemental voltage
(−0.114 V) must be applied to the cell in order to drive current. Similar to in MFC, HER
cathode over-potentials significantly increase the magnitude of applied potential (0.3 –
1.0 V in practice) required to produce hydrogen in an MEC [17, 18].
12
2.1.2 Anode Materials
Graphitic carbon based materials are most commonly used as anodes in both MFCs
and MECs. Graphite is advantageous because it is highly electrically conductive, does not
corrode at the operating potential and pH of the electroacitve biofilms, microbes attach
well to the material, and it can be made from waste biomass. High surface area and
porous materials such as graphite fiber brushes have been used to maximize area (4,600
m2/m3 reactor) for biofilm attachment and minimize substrate transport limitations [19].
Anodes can also be thermally or ammonia treated before inoculation to promote
attachment and decrease enrichment times [20, 21].
Figure 2-1: Diagram of microbial electrochemical technologies.
Organic(Carbon(
CO2(+(4H+(
e1(
e1( e1(
e1(
O2(+(4H+(
2H2O(
O2(+(4H+(
H2O(
4H+(
2H2(
e1( e1(
2H2O(+(O2(
2H2O2(
NO2(
½N2(
13
2.1.3 Cathode Materials
Oxygen reduction in MFCs occurs at the intersection of three phases (liquid, air and
solid) and must simultaneously prevent water leakage while allowing for the transport of
oxygen to the catalytic interface. MFC air breathing cathodes are constructed on a planar
current collector by applying a catalyst layer to the water facing side, and multiple
hydrophobic oxygen diffusion layers to the air facing side [2]. Cathode material cost and
performance are main limiting factors in MFC scale up and performance. To lower
catalyst costs, alternative non-noble catalysts have been explored with recent results
indicating that activated carbon can replace platinum as an oxygen reduction catalyst
without significant losses in power production [22, 23].
The performance of hydrogen evolution cathodes in MECs is highly dependent on the
activation potential loss (or over-potential) of the inert metal catalyst. When platinum, an
ideal HER catalyst with an over-potential near −0.1 V, is used the applied potential
required to produce hydrogen is as low as 0.3 V and the energy produced as hydrogen
can be nearly three times greater than the electricity invested to apply a cell potential
[17]. Inexpensive metal alloys such as stainless steel 304 can be used in MECs but
significantly higher potentials (0.6 – 1.0 V) are required to overcome the HER over-
potential (-0.7 V) [18, 24].
2.1.4 Membranes
Ion exchange membranes are often used to separate the anode and cathode chambers
of METs. In MFCs, anion, cation, and proton exchange membranes as well as filtration
membranes have been used to prevent oxygen intrusion into the anode [25, 26]. In
14
general, dual-chambered MFCs produce significantly less power than in single chamber
reactors because at neutral pH, and in low conductivity solutions such as wastewater,
membranes substantially increase internal resistance and lead to acidification of the
anode and alkalization of the cathode. Trans-membrane pH gradients and also incur
potential losses, estimated at 60 mV per pH unit difference across the membrane.
Negative membrane potentials can increase (0.4 – 1.0 V) the applied potential required to
produce hydrogen in MEC [27].
2.1.5 Internal Resistance
MET performance is related to factors that resist current production and ionic
transport. In MFCs power production (P) is a function of both cell voltage (V) and
external resistance (Rext) and maximum power is reached when the external resistance is
equal to the internal resistance of the cell (PMFC = IV = V2/Rint) [1]. When current (I) in
generated, internal resistances lead to cell voltage losses which reduces power production
in MFCs and increases the required applied potential in MECs. Potential losses can occur
in solution (Rsol), across a membrane separator (Rmem) if one is present in the system, and
at the electrodes (Ran, Rcat). Internal resistance of METs is composed of ohmic, kinetic,
and diffusional resistances [28]. Ohmic resistance is a measure of how quickly charge is
transported through solution and it is affected by the conductivity of the anode and
cathode electrolytes, and the distance between electrode. Kinetic resistance, also known
as charge transfer resistance (Rct), is a measure of how favorable the anode and cathode
redox reactions are under a given set of system conditions [29]. Diffusion (Rdiff)
15
resistances are caused by transport imbalances at the electrode reaction interphase [30]
and across the membrane [31].
Electrochemical methods can be used to characterize bulk resistance of the reactor
components (i.e. Ran, Rmem, and Rcat) and specific ohmic and capacitive resistances (i.e.
Rsol, Rct, and Rdiff). Internal resistance can be measured from the slope of a voltage versus
current plot (linear slope method) [32]. The bulk resistances of each of the reactor
components can be determined measuring electrode or trans-membrane potential against
reference electrodes (typically Ag/AgCl) during a linear polarization experiment.
Electrochemical impedance spectroscopy (EIS) is often used to determine the ohmic and
capacitive resistances of a MFC [33]. In EIS experiments, a sinusoidal wave of potential
is applied to the electrochemical cell over a range of frequencies to induce a phase shift in
current [30]. Using EIS is was determined that solution conductivity of wastewater (~1
mS/cm) dramatically limits power production by increasing both solution and diffusion
resistance at the cathode [28]. To optimize power production within the confines of non-
ideal wastewater characteristics, new reactor designs must be explored to minimize the
effect of low wastewater conductivity on power production.
2.2 Phosphorus Removal and Recovery from Wastewater
2.2.1 Introduction
Phosphorus is a vital and irreplaceable nutrient for all living organisms. Phosphate,
the soluble inorganic form of phosphorus, fulfills a diverse array of functions within
cellular organisms including energy storage and transfer (ATP, NADH), information
transfer (DNA, RNA), and structure (Phospholipid bilayer) [34].
16
Human activity has dramatically increased the mobilization of phosphorus over the
last half century [35]. Industrial and agricultural activities consume roughly 30 million
metric tonnes (MMt) of phosphorus each year, with the majority of P being utilized to
grow crops for food and biofuel production. Current rates of consumption (~18 MMt)
predominantly rely on mined phosphorus [36] and rapid rise in demand and price over the
last three decades have led some researchers to predict phosphorus reserves will be
depleted in the next century [35]. According to the U.S. Geological Survey, at the current
consumption rate, worldwide proven reserves of sedimentary phosphate rock would last
roughly 1,600 years. While the longevity of our reserves is debatable, the effect of
current phosphorus consumption on natural ecosystems is clearly unsustainable. Each
year, 25.3 MMT of P are lost to landfills and waterways [35], directly contributing to the
exponential rise in coastal eutrophication since 1960 [37]. According to a recent
watershed survey conducted by the U.S. EPA, 53% of rivers and streams, 67% of lakes,
reservoirs and ponds, 67% of bays and estuaries, and 86% of coastal shorelines are
considered impaired by pollution [38]. Nutrients are the third largest contributor to
watershed impairment [38].
Livestock and human waste effluents are major contributors to aquatic eutrophication.
Animal feeding operations in the United States generate 1.27 billion tonnes of animal
waste each year, which is 137 times the amount generated by the human population [39].
These waste streams contain 11 MMt of P, and could be used to offset 55% of global P
demand annually if the phosphorus could sustainably recovered during waste treatment
[35]. Unfortunately, despite efforts by the EPA, a recent court ruling vacated the
requirement that concentrated animal feeding operations (CAFOs) that intend to
17
discharge to watersheds apply for a national pollution discharge elimination system
(NPDES) permit and meet nutrient discharge limits [40]. If adequate technology is
developed to economically and sustainably treat animal wastes and recover nutrients, it is
possible that this decision could be reversed.
2.2.2 Phosphorus Management at Wastewater Treatment Plants
Municipal wastewater treatment plants offer the opportunity to capture significant
amounts of phosphorus (3 MMt yr-1) within an established infrastructure [41].
Phosphorus concentrations entering wastewater treatment plants are dilute (~9 mg-P/L)
and sustainable treatment systems must simultaneously achieve removal from solution
and concentration of phosphorus in a reusable form [42]. The most commonly employed
method of phosphorus removal from primary wastewater streams that fits both criteria is
biological phosphorus removal.
Biological phosphorus removal can occur from bacterial cell growth and enhanced
uptake for energy storage. Since the first observation of biological phosphorus uptake by
aerated sludge in the 1950s, biological nutrient removal has undergone extensive
microbiological investigation and process engineering, and it is now a widely used and
effective biotechnology for phosphorus removal from municipal wastewater [43]. The
enhanced biological phosphorus removal (EBPR) process relies on a group of bacterial
known as phosphate accumulating organisms (PAO) that are capable of absorbing and
storing P in excess of cell growth requirements [43].
The EBPR process requires a strictly anaerobic zone immediately followed by an
aerobic or anoxic zone. In the anaerobic zone, PAOs use stored polyphosphates to
18
generate adenosine triphosphate (ATP). ATP and absorbed acetate are consumed to
generate polyhydroxyalkanoates (PHAs) an energy storage molecule. The strictly
anaerobic phase is required to ensure the presence of VFAs from fermentation and
provide PAO an opportunity to outcompete other microbes for VFAs. Under
aerobic/anoxic conditions and in the absence of volatile fatty acids, the PAO utilize
stored PHB as a substrate to produce new biomass and restore pools of polyphosphates
and glycogen. Soluble P is absorbed from solution to restore poly-P reserves. If both
VFAs and an electron donor (O2, NO2/NO3) are present, then PAOs will make PHA and
phosphate will be released. The PAO Accumulibacter phosphatis has been shown to
dominate (~80%) the microbial ecology of EBPR sludge [44].
Clarified EBPR sludge contains 5 - 7% of phosphorus by mass and can be reused as
an agricultural fertilizer if risks associated with transmission of pathogens and heavy
metals are eliminated [45]. Thermophilic anaerobic digestion of EBPR sludge is a
common method of stabilization.
2.2.3 Phosphorus Precipitation from Nutrient Rich Side-streams
Aerobic biological wastewater treatment such as activated sludge and EBPR generate
excessive quantities of sludge. Sludge processing and disposal can account for up to 65%
of plant operating costs [46]. To stabilize and thicken sludge prior to disposal or land
application, solids are commonly anaerobically digested and the produced effluent
centrifuged. The remaining liquid often contains high levels of Ca2+, Mg2+, K+, PO43-, and
NH4+. During anaerobic digestion of EBPR sludge, PAOs release phosphate, and 40 -
80% of the phosphorus that enters the digester as sludge is solubilized and transferred to
19
the supernatant [47, 48]. The release of nutrients during digestion often leads to the
precipitation of calcium and magnesium phosphate salts within the digester as well as on
equipment surfaces. Scale formation on pumps and within conduit can lead to costly
equipment failure and conduit replacement [49-51]. The annual operational cost of
dealing with scale formation at wastewater treatment plants is a function of treatment
capacity and has been estimated to be around $10,000 per million gallons per day (MGD)
of dry weather capacity [52]. A number of mineral salts can form in nutrient rich digester
supernatant (Table 1) with the magnesium ammonium phosphate hexahydrate (struvite)
and amorphous calcium phosphate being the most prevalent.
Table 2.1: Molecular formulas and formation constants for phosphate and carbonate salts.
Common Name Formula pKso Ref.
Struvite MgNH4PO4·6H2O(s) 13.2 [53]
Newberyite MgHPO4·H2O(s) 18.2 [54]
Bobierrite Mg3(PO4)2·22H2O(s) 25.2 [54]
Calcium Hydrogen phosphate CaHPO4(s) 6.7 [55]
β-Tricalcium Phosphate (TCP) β-Ca3(PO4)2(s) 24.0 [55]
Hydroxyapatite (HAP) Ca5(PO4)3OH(s) 55.9 [55] Calcite CaCO3(s) 8.3 [55]
Heterogeneous solid precipitation occurs when the ionic concentration product
components exceeds the thermodynamic solubility limit for a given solid. The primary
growth potential of a salt has been defined as the ratio of the conditional ionic product
(IP) to the solubility limit. Saturation is reached when the ratio is ≥ 1. For struvite (2) and
calcium phosphate (3), the ratios are [53, 56]:
𝛽 = !"!!"
(2.4)
20
IPStruvite = (αMgγMgCT,Mg) (αMgγMgCT,Mg) (αPO4γPO4CT,PO4) (2.5)
IPTCP = (αCaγCaCT,Ca)3(αPO4γPO4CT,PO4) 2 (2.6)
IPHAP = (αCaγCaCT,Ca)5(αPO4γPO4CT,PO4) 3(αOHγOHCT,OH) (2.7)
where β is the supersaturation ratio, IP is the ionic product, α is the ionization fraction of
each component ion, γ is the activity coefficient, and CT is the total measure
concentration of the ion in solution.
The ionic product of struvite and calcium phosphate are both strongly affected by pH.
At equimolar concentrations, supersaturation of struvite occurs in the shared availability
regions of ammonium (pKa = 9.3) and ortho-phosphate. The availability of ortho-
phosphate increases as pH approaches 12.3 (pKa3 of phosphate) but minimum solubility
is reached at pH = 9.3 – 10.7 [51, 55]. In carbonate rich solutions (>10mM) such as
digestate, complexation of HCO3– and CO3
2– with calcium and magnesium increases the
solubility struvite and calcium phosphate salts [57, 58]. High concentrations of carbonate
have shown to lower the pKso of hydroxylapatite from 56 at pH = 7 to 44 at pH = 11 [57].
Phosphate salt crystal formation occurs in solution from primary nucleation (the
formation of new crystals) and secondary growth onto existing crystal surfaces [59, 60].
Secondary growth on the surface of granules is governed by mass transfer conditions and
surface reaction kinetics and has been modeled as a two-step process [59]:
!"!"= 𝑘!(𝛽 − 𝛽!) (2.8)
!"!"= 𝑘!𝛽!! (2.9)
where dm/dt is the crystal accumulation rate, m is moles of salt, t is reaction time (s), β is
the overall supersaturation and βi is the supersaturation at the granule surface-solution
21
interface, kd is the overall diffusion coefficient of the salt components, kr is the reaction
constant of the salt, and x is the surface reaction order.
To preventing equipment failure and recovering nutrients has led to development of
digestate and centrate treatment systems. Both physical and chemical methods have been
explored to induce supersaturation of phosphate salts. Precipitation techniques typically
involve raising solution pH to increase the availability of phosphate ions. Sodium
hydroxide (NaOH) is often used because it is highly soluble and can rapid increase the
solution pH of alkaline wastewater between 8.0 and 10.0 [49, 61-63]. Magnesium and
calcium bases (Mg(OH)2 and MgO) are attractive because, in addition to raising pH,
supplemental soluble cation further increases the IP of phosphate salts [54]. Stripping
carbonate alkalinity by aeration has also been shown to effectively raise pH and reduce
cationic complexation with carbonate [56, 64].
Figure 2.2: Examples of scale accumulation in pipes and crystallized phosphate salts recovered from a fluidized bed reactor.
22
The two most popular reactor designs employed for phosphate salt crystallization are
aeration basins and fluidized beds. Aeration basin crystallization systems are designed to
simultaneously raise pH by stripping alkalinity, promote secondary crystal growth with
upward turbulent mixing, and continuously suspend growing salt granules. Although
simple in design, aeration basins are energy intensive to operate and provide little control
of particle size. Fluidized bed reactors (FBRs) are tall and narrow columns designed to
generate up-flow velocities strong enough to vertically segregate a bed of granules
according to particle diameter. FBRs are often used for crystallized phosphate recovery
because the design creates an abundance of reaction surface area and solution turbulence
[60]. These conditions promote rapid molecular growth at the surface interface of the
granules essentially eliminating the induction time of nucleation [55]. As a result, 70 −
90% removal of soluble phosphate at low hydraulic retention times (1 − 4 h) are typical
for FBRs [60, 61, 65, 66]. Another major advantage of fluidized beds is that by
Figure 2.3: Diagram of a commercial scale fluidized bed reactor for struvite precipitation.
23
controlling up-flow velocity within the reactor phosphate crystals can grow to a desired
size (3 − 5 mm at full scale), settle and be harvested without shutting down the reactor
[66]. The settling velocity of a particle of known particle density is reached when the
gravitational force acting on the particle matches the drag force exerted by the viscous
liquid fluidizing the particle within the reactor:
𝜗! =!!(!!! !! )
!𝑔𝑅! (2.10)
where υs is the settling velocity (m/s), ρp is the particle density (kg/m3), R is the particle
radius (m), ρf is the fluid density (kg/m3), µ is the fluid dynamic viscosity (N-s/m2), and g
is gravitational acceleration (m/s2) [67].
2.2.4 Electrochemical Phosphate Salt Precipitation
Recently, electrochemical methods of inducing struvite precipitation have been
explored. The fundamental attraction to electrochemical methods tends to revolve around
replacing pH increases by hydroxide addition with consumption of protons via hydrogen
evolution [68, 69]. The counter electrode oxidation reaction was originally water
oxidation, and more recently metal magnesium passivation [70]. Tested electrochemical
systems have been operated with both electrodes in the same reactor chamber. In water
electrolysis, precipitation at cathode surface can occur due to a localized pH increase
caused by hydrogen evolution. In the Mg dissolution cell, a bulk phase pH increase was
created by the release of Mg2+ into solution from oxidation of Mg(0) at the anode, and the
consumption of H+ at the cathode. Both of these systems suffer from critical flaws: water
electrolysis required >3.0 V of applied voltage, making energy input prohibitively
24
expensive [68]. The electrolytic magnesium passivation system had a low energy input
(0.1 - 2.2 Wh/g-P), but the cost of magnesium electrode led to a struvite production cost
of $6/kg-struvite [70]. Most critically, the effect of cathode scale formation on the
sustainability of current generation, and thus pH increase, has not been explored with any
of the electrochemical struvite precipitation systems.
2.3 Reverse Electrodialysis
2.3.1 Introduction
When volumes of fresh and salt water mix, energy is released, the magnitude of
which is a function of the molar entropy difference between the concentrated and dilute
saline solutions [71]. For ideal solutions, the Gibb’s free energy of mixing (ΔGmix) can be
described as:
ΔGmix = ΔGM – (ΔGHC + ΔGLC) (2.11)
ΔGmix = -(nHC + nLC)TΔSM + (nHCTΔSHC + nHCTΔSLC) (2.12)
ΔS = -RΣxilnxi (2.13)
where subscript HC represents high concentrate solution, LC represents low concentrate
solution, M is the resulting mixed solution, n is moles, ΔS is the molar entropy, R is the
gas rate constant, T is temperature, and xi is the mole fraction of an ionic substance in
solution [72].
Mixing 1 m3 of seawater (~ 0.5 M NaCl) and 1 m3 of freshwater (~0.01 NaCl)
theoretically releases 1.4 MJ of entropic energy [72, 73]. Based on the global discharge
of river water (1.3 × 106 m3/s) into the ocean, roughly 2.8 TW of mixing energy is
released worldwide [74, 75]. Discharge of treated wastewater into the sea releases
25
another 18.5 GW of energy [76]. Recent research has also explored using ammonium
bicarbonate (NH4HCO3), a thermolytic salt that decomposes at low temperatures (40 – 60
°C), to generate salinity gradient from thermal energy such as industrial waste heat [77,
78]. Each year in the United States over 204 GW of energy is lost as heat in the industrial
sector [79]. One advantage of engineered salinity gradients is that the salinity ratios are
not predetermined by the natural bodies of water and can be increased up to the solubility
of the mixing salt (2 M for AmB). Mixing 1 m3 of concentrated AmB (2.0 M NaCl) with
1 m3 of freshwater (~0.01 AmB) theoretically releases 5.4 MJ (1.5 kWh) of entropic
energy.
Engineers have designed systems to control salinity gradient mixing and convert free
energy into electrical energy. Research and development efforts have focused on two
membrane-based systems: pressure retarded osmosis (PRO) and reverse electrodialysis
(RED) [76, 80]. PRO makes used of the osmotic pressure difference between saline and
fresh solutions to drive water flux across a thin uncharged membrane. As water transports
across the membrane, from dilute to concentrated, volumetric displacement can perform
work on a turbine and generate electricity [81]. RED systems generate electricity by
separating salt and fresh water streams with ion-selective membranes to control ionic
mixing, and convert ionic flux into electrical current. At the time of this work, AmB
solutions have only been used to generate salinity gradients for use in PRO [82] and have
not been used to generate power with RED based systems. Since one of the key
objectives of this dissertation is to recovery heat using AmB solutions within a microbial
RED cell, the material in this literature review will focus on the fundamentals of RED
power generation.
26
When an ion-selective barrier separates solutions of different ionic concentration, an
electrochemical junction potential (Δϕ) is observed. The potential produced by a HC/LC
junction, or cell pair, is related to the activity difference between the two solutions and
can be estimated with the Nerst equation,
Δϕ = 2(RgT/F) ln(γi,HCxi,HC/γi,LCxi,LC) = 2(RgT/F) ln(ai,HC/ai,LC) (2.14)
where γi represents the mole fraction based activity coefficient and ai is ion activity. The
theoretical potential for a cell pair of 0.5 M (seawater) and 0.01 NaCl (river water) is 220
mV [83]. The theoretical potential for a cell pair of 2.0 M and 0.01 AmB is 240 mV. A
fraction of the potential based on bulk phase concentration difference is lost to membrane
non-idealities and concentration polarization at the membrane interface, leading to
observed junction potentials that ranged from 0.1 – 0.2 V per cell pair [84, 85].
To produce power using RED, multiple cell pairs are constructed in series to drive
current generation at redox electrodes placed at either end of the membrane stack. At the
anode, an oxidation reaction releases electrons to an external circuit and generates a net
Figure 2.4: Diagram of a reverse electrodialysis reactor.
CEM$ AEM$HC#
Stream#LC#
Stream#
A,#
A,#
A,#
A,#
A,#
C+#
C+#
C+#
C+#
A,#
C+#
A,#
A,#
A,#
A,#
A,#
C+#
C+#
C+#
C+#
A,#
C+#
C+#
C+#
C+#A,#
A,#
A,#
2H2O$
O2$+$4H+$ 4H2O$
4OH+$+$2H2$
e+$ e+$
e+$e+$
27
positive charge increase in the anode chamber. Electrons flow from anode to cathode
through an external circuit. At the cathode a reduction reaction occurs, resulting in a net
decrease in positive charge in the cathode chamber. Solution charge imbalance allows
ionic flux through the selective membranes of the RED stack. The membranes are
arranged to direct anionic flux toward the anode and cationic flux to the cathode to
maintain electro-neutrality within the system. As current flows through the external
circuit, ionic mixing between the HC and LC chambers increases, decreasing the additive
junction potential of the stack [80, 86].
2.3.2 Reverse Electrodialysis Power Production
The power produced by the RED is dependent on the amount of current passing
between the electrodes and the net cell voltage. Maximum power is produced when the
resistance the external circuit (Rext) is equal to the internal resistance of the system (Rint).
Gross power production can be estimated from the open circuit voltage (OCV) and
internal resistance of an electrochemical cell [84].
W = I2Rload = IV = (Vo)2Rext/(Rext + Rint)2 (2.15)
Wmax = (Vo)2Rext/4Rint (2.16)
The OCV of a stack in dependent on several factors including the salinity ratio the
salt solutions, number of salinity gradient cell pairs, membrane selectivity, fluid dynamic
within the stack, and the redox chemistry of the electrode pair. Internal resistance of the
cell is dependent on membrane properties fluid flow geometry, solution characteristics
and the kinetics of the redox catalysis electrodes (Rint = Rstack + Relectrodes) [87].
28
2.3.3 Membrane Resistance
In a RED system, electrical resistance arises from membrane properties that contribute to
ionic transport resistance. Ion-exchange membranes are relatively water impermeable
barriers that contain fixed charge groups that allow for the passage or either cations or
anions [88]. Anion exchange membranes typically contain fixed quaternary ammonium
groups that interact with anions in solution. Cation exchange membranes employ fixed
sulfonate groups that attract and transport soluble cations across a potential gradient.
Ionic transport in ion-exchange membranes is a function of material properties such as
conductivity, thickness, and the projected area for ionic transport.
The ion conductivity of a membrane, commonly referred to as the ion-exchange
capacity (IEC), determines how quickly ions diffuse through the interface and is
dependent on the density of fixed charged groups attached to the support polymer matrix
[89]. In theory, membrane conductivity and resistance have an inverse relationship: as the
number of fixed charges increases the faster counter-ions can pass through the interface.
However, membrane conductivity must be optimized to prevent membrane swelling that
occurs when soluble counter-ions and water molecules form hydrations shells around the
fixed charge groups, causing the membrane to swell [88]. Membrane swelling increases
the ion-exchange interface thickness, which can in turn increases ionic flux resistance.
Membrane geometry, specifically thickness and area, also greatly contribute to RED
stack resistance. Reported membrane resistance values are typically normalized to area
(i.e. Ω-cm2) indicating that resistance decreases as the area for ionic transport increases
[90].
29
2.3.4 Solution Resistance
Solution resistance can be calculated from solution conductivity and flow
chamber geometry (Rsol = K d/A). Because of low conductivity of freshwater (≤1
mS/cm), solution resistance in the low concentrate chambers severely limit power
production. The highest reported power production (2.2 W/m2) was reported in a study
that focused on reducing internal resistance by decreasing the solution flow chamber
thickness [91]. A thickness of 100 µm was determined to be optimal because are lower
thickness (60 µm), pressure drop within the stack increased substantially, which increased
the pumping energy input to the system.
Solution conductivity can also have a dramatic effect of the electrical resistance of
membrane interfaces by increasing the transport boundary layer thickness. As current
passes through an ion-exchange membrane, the majority of the ionic current will be due
to counter-ion transport, creating a localized counter-ion gradient at the membrane
interface that acts in opposition to the bulk concentration salinity gradient in a RED. This
transport phenomenon is known as concentration polarization and the thickness of this
ion-specific gradient is known as the transport boundary layer [92]. It was recently shown
that under stagnant conditions, observed membrane resistance increases from 1 – 3 Ω-
cm2 in 0.5 M NaCl to 10 – 20 Ω-cm2 for AEMs and 20 – 50 Ω-cm2 for CEM in fresh
water (0.017 M NaCl) [93]. The boundary layer thickness, and resistance, in dilute
electrolytes can be significantly reduced with hydrodynamic turbulence such as mixing
[94] or increased solution flow rate at the interface [93].
30
2.3.5 Spacers
Spacers are used to prevent membrane deformation and create mixing within a flow
path. Spacers add to internal resistance by blocking active membrane surface area, a
phenomenon commonly referred to as spacer shadowing effect. In one study, a spacer
was constructed from perforated anion and cation exchange membranes that were
sandwiched together in the RED solution flow paths. Although this is not an economical
solution to spacer shadowing, power production was shown to increase by ~300% [95]
Profiled membranes that create linear flow paths when positioned side by side have also
been used to eliminate the use of spacers [96]. While this approach eliminated spacer
resistance, the linear flow paths lead to an increase in boundary layer resistance and
minimally affected power density.
2.3.6 Redox Electrode Pairs
Since electrodes covert the ionic current of RED stacks to electrical current, proper
consideration must be paid to the redox couples chosen for current generation. An
effective redox couple must be sustainable, and minimize energy loss due to reaction
kinetics (over-potentials). In the last thirty years, the evolution of RED electrodes has
transitioned from reversible reactive electrodes [71, 97], to inert electrodes with reactive
or capacitive electrolytes [85, 96, 98, 99], to most recently, biologically active electrodes
[100, 101].
Reversible redox electrodes utilize identical reactive metal electrodes and an
electrode rinse. At the anode the current is generated by metal corrosion (i.e. Zn → Zn2+
31
+ 2 e-). The electrode rinse, containing a background concentration of NaCl) carries the
dissolved metal ions from the anode to cathode, where the metal ion is reduced (Zn2+ + 2
e- → Zn). The benefit of this system is that there is no net electrode resistance and current
could be generated with only a few cell pairs. However, reversible electrode pair design
is not sustainable because of metal loss at the anode and accumulation at the cathode
(approximately 2 mm of electrode thickness per week at 50 A/m2) [71].
To avoid electrode material transport, researchers have explored coupling inert
electrodes (e.g. graphite or Pt/Ir) with homogeneous redox active iron based electrolytes.
The soluble redox couples tested include ferric chloride [99] and hexacyanoferrate
([Fe(CN)6]3-/[Fe(CN)6]4-) [91]. In both systems, ferric iron donates one electron to the
anode and receives one electron at the cathode. The benefit of this system is the same as
the reversible electrode couple in that there is no potential loss as the electrodes. The
drawbacks of this system are iron fouling of the ion-exchange membranes, iron
precipitation at the cathode surface and the possibility of cyanide or chlorine gas
formation at electrodes [99, 100].
Water splitting electrodes has been used to avoid membrane fouling and toxic gas
formation. At the anode, water is oxidized to oxygen or hydroxyl radicals. At the cathode,
water is reduced to form hydrogen gas. The theoretical minimum potential for this
reaction is 1.24 V but due to catalyst inefficiency, the net resistance to current generation,
nearly 2.5 V is lost to electrode activation over-potentials [85].
Recently, RED membrane stacks have been paired with MFC electrodes [100]. MFCs
contain anodes where microbes grow and release electrons, and air cathodes, where
electrons are consumed in the oxygen reduction reaction [102]. By coupling a RED stack
32
with spontaneous MFC electrode kinetics (i.e., oxidation of organic matter at the anode
and oxygen reduction at the cathode), significant energy losses which occur with water
electrolysis (up to 2.5 V in cell voltage loss in a 25 membrane stack due to electrode
over-potentials) are prevented [85]. System damage associated with the hexacyanoferrate
redox couple ([Fe(CN)6]3-/[Fe(CN)6]4-) can also be avoided [72, 86, 103-105]. The
spontaneous nature of MRC current generation also allows salinity-gradient energy
generation with much fewer membranes (5 membrane pairs compared to 25 – 50) [100]
representing a substantial reduction in the membrane associated capitol costs ($100 –
200/m2) [76].
33
2.4 Literature Cited
1. Logan, B.E., et al., Microbial fuel cells: methodology and technology. Environmental
Science & Technology, 2006. 40(17): p. 5181-5192.
2. Liu, H. and B.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, 2004. 38(14): p. 4040-4046.
3. Liu, H., S. Grot, and B.E. Logan, Electrochemically assisted microbial production of
hydrogen from acetate. Environ Sci Technol, 2005. 39(11): p. 4317-4320.
4. ter Heijne, A., et al., Ferric iron reduction as an alternative for platinum-based cathodes
in microbial fuel cells. Environmental Science & Technology, 2006. 40: p. 5200-5205.
5. Heijne, A.T., et al., Copper Recovery Combined with Electricity Production in a
Microbial Fuel Cell. Environmental Science & Technology, 2010. 44(11): p. 4376-4381.
6. Rabaey, K., et al., High Current Generation Coupled to Caustic Production Using a
Lamellar Bioelectrochemical System. Environmental Science & Technology, 2010.
44(11): p. 4315-4321.
7. Aulenta, F., et al., Electron transfer from a solid-state electrode assisted by methyl
viologen sustains efficient microbial reductive dechlorination of TCE. Environmental
Science & Technology, 2007. 41(7): p. 2554-2559.
8. Nevin, K.P., et al., Electrosynthesis of organic compounds from carbon dioxide is
catalyzed by a diversity of acetogenic microorganisms. Applied and Environmental
Microbiology, 2011. 77(9): p. 2882-2886.
9. Rabaey, K. and R.A. Rozendal, Microbial electrosynthesis- revisiting the electrical route
for microbial production. Nat Rev Micro, 2010. 8(10): p. 706-716.
34
10. Cusick, R.D., P.D. Kiely, and B.E. Logan, A monetary comparison of energy recovered
from microbial fuel cells and microbial electrolysis cells fed winery or domestic
wastewaters. International Journal of Hydrogen Energy, 2010. 35(17): p. 8855-8861.
11. Kiely, P.D., et al., Anode microbial communities produced by changing from microbial
fuel cell to microbial electrolysis cell operation using two different wastewaters.
Bioresource Technology, 2011. 102(1): p. 388-394.
12. Call, D.F., R.C. Wagner, and B.E. Logan, Hydrogen production by Geobacter species
and a mixed consortium in a microbial electrolysis cell. Applied and Environmental
Microbiology, 2009. 75(24): p. 7579-7587.
13. Malvankar, N.S., et al., Tunable metallic-like conductivity in microbial nanowire
networks. nature nanotechnology, 2011. 6(9): p. 573-579.
14. Cheng, S., H. Liu, and B.E. Logan, Increased power generation in a continuous flow
MFC with advective flow through the porous anode and reduced electrode spacing.
Environmental Science & Technology, 2006. 40: p. 2426-2432.
15. Logan, B.E., Exoelectrogenic bacteria that power microbial fuel cells. Nature Reviews
Microbiology, 2009. 7(5): p. 375-381.
16. Rismani-Yazdi, H., et al., Cathodic limitations in microbial fuel cells: An overview.
Journal of Power Sources, 2008. 180: p. 683–694.
17. Call, D. and B.E. Logan, Hydrogen production in a single chamber microbial electrolysis
cell (MEC) lacking a membrane. Environmental Science & Technology, 2008. 42(9): p.
3401-3406.
18. Call, D., M.D. Merrill, and B.E. Logan, High surface area stainless steel brushes as
cathodes in microbial electrolysis cells (MECs). Environmental Science & Technology,
2009. 43(6): p. 2179-2183.
35
19. Logan, B.E., et al., Graphite fiber brush anodes for increased power production in air-
cathode microbial fuel cells. Environmental Science & Technology, 2007. 41(9): p.
3341-3346.
20. Cheng, S. and B.E. Logan, Ammonia treatment of carbon cloth anodes to enhance power
generation of microbial fuel cells. Electrochem Commun, 2007. 9(3): p. 492-496.
21. Wang, X., et al., Use of Carbon Mesh Anodes and the Effect of Different Pretreatment
Methods on Power Production in Microbial Fuel Cells. Environmental Science &
Technology, 2009. 43(17): p. 6870-6874.
22. Watson, V.J., C. Nieto Delgado, and B.E. Logan, Influence of chemical and physical
properties of activated carbon powders on oxygen reduction and microbial fuel cell
performance. Environmental Science & Technology, 2013.
23. Zhang, F., D. Pant, and B.E. Logan, Long-term performance of activated carbon air
cathodes with different diffusion layer porosities in microbial fuel cells. Biosensors and
Bioelectronics, 2011. 30(1): p. 49-55.
24. Zhang, Y., M.D. Merrill, and B.E. Logan, The use and optimization of stainless steel
mesh cathodes in microbial electrolysis cells. International Journal of Hydrogen Energy,
2010. 35(21): p. 12020-12028.
25. Kim, J.R., et al., Power generation using different cation, anion and ultrafiltration
membranes in microbial fuel cells. Environmental Science & Technology, 2007. 41(3): p.
1004-1009.
26. Harnisch, F., U. Schröder, and F. Scholz, The suitability of monopolar and bipolar ion
exchange membranes as separators for biological fuel cells. Environmental Science &
Technology, 2008. 42: p. 1740–1746.
36
27. Rozendal, R.A., et al., Effect of the type of ion exchange membrane on performance, ion
transport, and pH in biocatalyzed electrolysis of wastewater. Water Science and
Technology, 2008. 57(11): p. 1757-1762.
28. Hutchinson, A.J., J.C. Tokash, and B.E. Logan, Analysis of carbon fiber brush loading in
anodes on startup and performance of microbial fuel cells. Journal of Power Sources,
2011. 196(22): p. 9213-9219.
29. Manohar, A.K. and F. Mansfeld, The internal resistance of a microbial fuel cell and its
dependence on cell design and operating conditions. Electrochimica Acta, 2009. 54(6): p.
1664-1670.
30. Bard, A.J. and L.R. Faulkner, Electrochemical Methods: Fundamentals and Applications.
2nd ed2001, New York: John Wiley & Sons.
31. Ji, E., et al., Interface resistances of anion exchange membranes in microbial fuel cells
with low ionic strength. Biosensors and Bioelectronics, 2011. 26(7): p. 3266-3271.
32. Logan, B.E., Microbial fuel cells2008, Hoboken, NJ: John Wiley & Sons, Inc. 300.
33. He, Z. and F. Mansfeld, Exploring the use of electrochemical impedance spectroscopy
(EIS) in microbial fuel cell studies. Energy & Environmental Science, 2009. 2(2): p. 215-
219.
34. Madigan, M.T., et al., Brock Biology of Microorganisms. Eleventh ed2009, Upper Saddle
River, NJ: Pearson Eductation Inc.
35. Cordell, D., J.O. Drangert, and S. White, The story of phosphorus: Global food security
and food for thought. Global Environmental Change, 2009. 19(2): p. 292-305.
36. Villalba, G., et al., Global Phosphorus Flows in the Industrial Economy From a
Production Perspective. Journal of Industrial Ecology, 2008. 12(4): p. 557-569.
37. Diaz, R.J. and R. Rosenberg, Spreading dead zones and consequences for marine
ecosystems. Science, 2008. 321(5891): p. 926-929.
37
38. EPA, U.S. Watershed Assesment, Tracking & Environmental Results. 2013 [cited 2013
5/17/2013]; Available from: ofmpub.epa.gov/waters10/attains_nation_cy.control.
39. USDA, Manure Management for Water Quality: Costs to Animal Feeding Operations of
Applying Manure Nutrients to Land, E.R. Service, Editor 2003.
40. EPA, U.S., National Pollutant Discharge Elimination System Permit Regulation for
Concentrated Animal Feeding Operations: Removal of Vacated Elements in Response to
2011 Court Decision, 2012: The Federal Register. p. 44494 -44497.
41. Yuan, Z., S. Pratt, and D.J. Batstone, Phosphorus recovery from wastewater through
microbial processes. Current Opinion in Biotechnology, 2012.
42. Cornel, P. and C. Schaum, Phosphorus recovery from wastewater: needs, technologies
and costs. Water science and technology, 2009. 59(6): p. 1069.
43. van Loosdrecht, M.C.M., et al., Biological phosphate removal processes. Applied
Microbiology and Biotechnology, 1997. 48(3): p. 289-296.
44. Martín, H.G., et al., Metagenomic analysis of two enhanced biological phosphorus
removal (EBPR) sludge communities. Nature biotechnology, 2006. 24(10): p. 1263-1269.
45. Tchobanoglous, G., F.L. Burton, and H.D. Stensel, Wastewater Engineering Treatment
and Reuse (4th edn.). Metcalf & Eddy Inc, 2003, McGraw-Hill, New York.
46. Pérez-Elvira, S., P.N. Diez, and F. Fdz-Polanco, Sludge minimisation technologies.
Reviews in Environmental Science and Bio/Technology, 2006. 5(4): p. 375-398.
47. Jardin, N. and H. Popel, Phosphate release of sludges from enhanced biological P-
removal during digestion. Water science and technology, 1994. 30(6): p. 281-292.
48. Liao, P., D.S. Mavinic, and F.A. Koch, Release of phosphorus from biological nutrient
removal sludges: A study of sludge pretreatment methods to optimize phosphorus release
for subsequent recovery purposes. Journal of Environmental Engineering and Science,
2003. 2(5): p. 369-381.
38
49. Ohlinger, K.N., T.M. Young, and E.D. Schroeder, Kinetics effects on preferential struvite
accumulation in wastewater. Journal of Environmental Engineering, 1999. 125: p. 730.
50. Ohlinger, K.N., T.M. Young, and E.D. Schroeder, Postdigestion struvite precipitation
using a fluidized bed reactor. Journal of Environmental Engineering, 2000. 126: p. 361.
51. Doyle, J.D. and S.A. Parsons, Struvite formation, control and recovery. Water Research,
2002. 36(16): p. 3925-3940.
52. Neethling, J. and M. Benisch, Struvite control through process and facility design as well
as operation strategy. Water science and technology: a journal of the International
Association on Water Pollution Research, 2004. 49(2): p. 191.
53. Ohlinger, K.N., T.M. Young, and E.D. Schroeder, Predicting struvite formation in
digestion. Water Research, 1998. 32(12): p. 3607-3614.
54. Dempsey, B.A. Removal and reuse of ammonia and phosphate by precipitation of
struvite. in PROCEEDINGS OF THE INDUSTRIAL WASTE CONFERENCE-PURDUE
UNIVERSITY. 1998. MOSBY YEAR BOOK EUROPE LTD.
55. Snoeyink, V.L. and D. Jenkins, Water chemistry1980: John Wiley.
56. Battistoni, P., et al., Phosphorus removal from a real anaerobic supernatant by struvite
crystallization. Water Research, 2001. 35(9): p. 2167-2178.
57. Ferguson, J.F. and P.L. McCarty, Effects of carbonate and magnesium on calcium
phosphate precipitation. Environmental Science & Technology, 1971. 5(6): p. 534-540.
58. Cao, X. and W. Harris, Carbonate and Magnesium Interactive Effect on Calcium
Phosphate Precipitation. Environmental Science & Technology, 2007. 42(2): p. 436-442.
59. Bhuiyan, M.I.H., D. Mavinic, and R. Beckie, Nucleation and growth kinetics of struvite
in a fluidized bed reactor. Journal of Crystal Growth, 2008. 310(6): p. 1187-1194.
39
60. Seckler, M.M., O.S.L. Bruinsma, and G.M. Van Rosmalen, Phosphate removal in a
fluidized bed--I. Identification of physical processes. Water Research, 1996. 30(7): p.
1585-1588.
61. Bhuiyan, M., D. Mavinic, and F. Koch, Phosphorus recovery from wastewater through
struvite formation in fluidized bed reactors: a sustainable approach. Water Science &
Technology, 2008. 57(2): p. 175-181.
62. Le Corre, K.S., et al., Struvite crystallisation and recovery using a stainless steel
structure as a seed material. Water Research, 2007. 41(11): p. 2449-2456.
63. Le Corre, K.S., et al., Impact of calcium on struvite crystal size, shape and purity. Journal
of Crystal Growth, 2005. 283(3-4): p. 514-522.
64. Moerman, W., et al., Phosphate removal in agro-industry: Pilot- and full-scale
operational considerations of struvite crystallization. Water Research, 2009. 43(7): p.
1887-1892.
65. Seckler, M.M., O.S.L. Bruinsma, and G.M. Van Rosmalen, Calcium phosphate
precipitation in a fluidized bed in relation to process conditions: A black box approach.
Water Research, 1996. 30(7): p. 1677-1685.
66. Forrest, A., et al., Optimizing struvite production for phosphate recovery in WWTP.
Journal of Environmental Engineering, 2008. 134: p. 395.
67. Munson, B.R., et al., Fundamentals of fluid mechanics. Vol. 3. 1998: Wiley New York.
68. Moussa, S.B., et al., Electrochemical precipitation of struvite. Electrochemical and Solid-
State Letters, 2006. 9: p. C97.
69. Wang, C.-C., et al., Formation of pure struvite at neutral pH by electrochemical
deposition. Chemical Engineering Journal, 2010. 159(1): p. 280-283.
70. Hug, A. and K.M. Udert, Struvite precipitation from urine with electrochemical
magnesium dosage. Water Research, 2012.
40
71. Pattle, R., Production of electric power by mixing fresh and salt water in the
hydroelectric pile. 1954.
72. Post, J.W., H.V.M. Hamelers, and C.J.N. Buisman, Energy Recovery from Controlled
Mixing Salt and Fresh Water with a Reverse Electrodialysis System. Environmental
Science & Technology, 2008. 42(15): p. 5785-5790.
73. Post, J., et al., Towards implementation of reverse electrodialysis for power generation
from salinity gradients. Desalination and Water Treatment, 2010. 16(1-3): p. 182-193.
74. Veerman, J., Reverse Electrodialysis: Design and Optimization by Modeling and
Experimentation2010: University of Groningen.
75. Wick, G.L., Power from salinity gradients. Energy, 1978. 3(1): p. 95-100.
76. Ramon, G.Z., B.J. Feinberg, and E.M.V. Hoek, Membrane-based production of salinity-
gradient power. Energy & Environmental Science, 2011.
77. McCutcheon, J.R., R.L. McGinnis, and M. Elimelech, A novel ammonia-carbon dioxide
forward (direct) osmosis desalination process. Desalination, 2005. 174: p. 1-11.
78. McGinnis, R.L. and M. Elimelech, Energy requirements of ammoina-carbon dioxide
forward osmosis desalination. Desalination, 2007. 207(370-382).
79. DOE, U.S., Waste Heat Recovery: Technology and Opportunities in U.S. Industry, I.T.P.
U.S. Department of Energy, Editor 2008.
80. Post, J.W., et al., Salinity-gradient power: Evaluation of pressure-retarded osmosis and
reverse electrodialysis. Journal of Membrane Science, 2007. 288(1-2): p. 218-230.
81. Lee, K., R. Baker, and H. Lonsdale, Membranes for power generation by pressure-
retarded osmosis. Journal of Membrane Science, 1981. 8(2): p. 141-171.
82. McGinnis, R.L., J.R. McCutcheon, and M. Elimelech, A novel ammonia-carbon dioxide
osmotic heat engine for power generation. Journal of Membrane Science, 2007. 305: p.
13-19.
41
83. Post, J.W., Blue Energy: electricity production from salinity gradients by reverse
electrodialysis2009: Wageningen Universiteit.
84. Długołȩcki, P., et al., Practical Potential of Reverse Electrodialysis As Process for
Sustainable Energy Generation. Environmental Science & Technology, 2009. 43(17): p.
6888-6894.
85. Veerman, J., et al., Reverse electrodialysis: Comparison of six commercial membrane
pairs on the thermodynamic efficiency and power density. Journal of Membrane Science,
2009. 343(1-2): p. 7-15.
86. Lacey, R., Energy by reverse electrodialysis. Ocean engineering, 1980. 7: p. 1-47.
87. Veerman, J., et al., Reverse electrodialysis: A validated process model for design and
optimization. Chemical Engineering Journal, 2011. 166(1): p. 256-268.
88. Strathmann, H., Ion-exchange membrane separation processes. Vol. 9. 2004: Access
Online via Elsevier.
89. Xu, T., Ion exchange membranes: state of their development and perspective. Journal of
Membrane Science, 2005. 263(1): p. 1-29.
90. Dlugolecki, P., et al., Current status of ion exchange membranes for power generation
from salinity gradients. Journal of Membrane Science, 2008. 319(1-2): p. 214-222.
91. Vermaas, D.A., M. Saakes, and K. Nijmeijer, Doubled power density from salinity
gradients at reduced intermembrane distance. Environmental Science & Technology,
2011. 45(16): p. 7089-7095.
92. Choi, J.-H., J.-S. Park, and S.-H. Moon, Direct measurement of concentration
distribution within the boundary layer of an ion-exchange membrane. Journal of Colloid
and Interface Science, 2002. 251(2): p. 311-317.
42
93. Długołęcki, P., et al., On the resistances of membrane, diffusion boundary layer and
double layer in ion exchange membrane transport. Journal of Membrane Science, 2010.
349(1): p. 369-379.
94. Park, J.-S., et al., An electrical impedance spectroscopic (EIS) study on transport
characteristics of ion-exchange membrane systems. Journal of Colloid and Interface
Science, 2006. 300(2): p. 655-662.
95. Dlugolecki, P., et al., Ion conductive spacers for increased power generation in reverse
electrodialysis. Journal of Membrane Science, 2010. 347(1-2): p. 101-107.
96. Vermaas, D.A., M. Saakes, and K. Nijmeijer, Power generation using profiled
membranes in reverse electrodialysis. Journal of Membrane Science, 2011. 385–386(0):
p. 234-242.
97. Audinos, R., Electrodialyse inverse. Etude de l'energie electrique obtenue a partir de
deux solutions de salinites differentes. Journal of Power Sources, 1983. 10(3): p. 203-
217.
98. Vermaas, D.A., et al., Clean energy generation using capacitive electrodes in reverse
electrodialysis. Energy & Environmental Science, 2013.
99. Veerman, J., et al., Reverse electrodialysis: evaluation of suitable electrode systems.
Journal of Applied Electrochemistry, 2010. 40(8): p. 1461-1474.
100. Kim, Y. and B.E. Logan, Microbial Reverse Electrodialysis Cells for Synergistically
Enhanced Power Production. Environmental Science & Technology, 2011. 45(13): p.
5834-5839.
101. Kim, Y. and B.E. Logan, Hydrogen production from inexhaustible supplies of fresh and
salt water using microbial reverse-electrodialysis electrolysis cells. Proceedings of the
National Academy of Sciences, 2011.
43
102. Liu, H., R. Ramnarayanan, and B.E. Logan, Production of electricity during wastewater
treatment using a single chamber microbial fuel cell. Environmental Science &
Technology, 2004. 38(7): p. 2281-2285.
103. Veerman, J., et al., Reverse electrodialysis: Performance of a stack with 50 cells on the
mixing of sea and river water. Journal of Membrane Science, 2009. 327(1-2): p. 136-144.
104. Vermaas, D.A., M. Saakes, and K. Nijmeijer, Doubled power density from salinity
gradients at reduced inter-membrane distance. Environmental Science & Technology,
2011. 45: p. 7089-7095.
105. Vermaas, D.A., M. Saakes, and K. Nijmeijer, Power generation using profiled
membranes in reverse electrodialysis. Journal of Membrane Science, 2011.
Chapter 3
Energy Efficient Phosphate Recovery as Struvite Within a Single
Chamber Microbial Electrolysis Cell
3.1 Abstract
An energy efficient method of concurrent hydrogen gas and struvite (MgNH4PO4·6H2O)
production was investigated based on bioelectrochemically driven struvite crystallization
at the cathode of a single chamber microbial electrolysis struvite-precipitation cell
(MESC). The MESC cathodes were either stainless steel 304 mesh or flat plates.
Phosphate removal ranged from 20 – 40%, with higher removals obtained using mesh
cathodes than with flat plates. Cathode accumulated crystals were verified as struvite
using a scanning electron microscope capable of energy dispersive spectroscopy (SEM-
EDS). Crystal accumulation did not affect the rate of hydrogen production in struvite
reactors. The rate of struvite crystallization (g/m2-hr) and hydrogen production (m3/m3-d)
were shown to be dependent on applied voltage and cathode material. Overall energy
efficiencies (substrate and electricity) were high (73 ± 4%) and not dependant on applied
voltage. These results show that MESCs may be useful both as a method for hydrogen
gas production and struvite production.
45
3.2 Introduction
Phosphorous is an essential nutrient for all forms of life, but it exists at low
concentrations in the earth’s crust. Modern agricultural practices primarily rely on
synthetic fertilizers containing mined phosphate rock to increase crop yields. Extensive
mining of phosphorus to meet global demands has led to speculation that known global
phosphate deposits may not last through the 21st century [1], with peak in phosphorus
production projected to occur in 2030 [2]. As worldwide reserves diminish, renewable
methods of phosphate production must be developed to help stabilize global food
production. One renewable source of phosphate is magnesium ammonium phosphate
hexahydrate (MgNH4PO4·6H2O), or struvite, which is known to be an excellent slow
release fertilizer [3, 4].
Struvite can be produced from crystallization of inorganics in nutrient rich animal and
domestic wastewaters. Crystal precipitation occurs when concentrations of Mg2+, NH4+,
and PO43– exceed the solubility limit for struvite formation. Concentrations of the three
struvite components increase during anaerobic digestion of biosolids, specifically cell
lysis (magnesium and phosphate) and protein degradation (ammonia) [5]. The
concentration of the specific chemical species needed to form struvite is highly pH
dependent, resulting in a struvite solubility minimum near pH = 10 [5]. However,
effective struvite crystallization has been observed at pH > 8 [6].
Controlled struvite recovery from wastewater can be achieved by several approaches:
chemical addition, carbon dioxide stripping or electrolysis. Most struvite recovery studies
have focused on increasing solution pH via chemical base addition (NaOH, Mg(OH)2,
Ca(OH)2) and stripping carbon dioxide through aeration [5, 7, 8]. While these approaches
46
are effective, operational costs associated with continuous chemical addition or blower
operation are high ranging from $140-460/tonne of struvite (as compared to $40-50/tonne
of mined phosphate rock) [6]. Chemical base addition can account for up to 97% of
struvite precipitation costs [9]. If operation costs were reduced, sale of recovered struvite
could replace a monetary sink with a revenue stream.
Struvite can also be crystallized at the cathode of a water electrolysis cell where the
consumption of protons (via hydrogen evolution) results in a localized pH increase [10].
The main disadvantage of this process is the high energy cost for producing the potential
needed to split water at the anode (1.25 V in theory, or more than 1.8 V in practice).
Electrochemical precipitation studies have been conducted in the presence of turbulent
mixing (which in itself can cause precipitation), and have not investigated phosphate
removal, hydrogen production or energy efficiency, or addressed whether precipitation is
possible in a solution with a carbonate concentration similar to anaerobic digestion
effluent.
In order to make electrochemical struvite precipitation more energy efficient, we
examined production of struvite in a microbial electrolysis cell (MEC). In an MEC,
microorganisms convert organic and inorganic matter into electrical current at a
significantly lower potential (minimum of ~0.2 V when bacteria used) than that needed
for splitting water [11]. In addition, no precious metals are needed for the MEC,
compared water electrolyzers, which utilize noble metal catalysts on the anode. Recovery
of the hydrogen gas can also offset energy costs, as electrical energy efficiencies in
MECs have reached 400% (a ratio of the energy in the hydrogen gas produced, to the
electrical energy needed) [11]. Thus, the recovery and utilization of the hydrogen gas for
47
sustaining the process could possibly reduce energy costs at wastewater treatment plants
with combined heat and power systems. It has also been shown that seawater, in
combination with a freshwater source (such as treated wastewater) can be used to provide
the needed potential using a reverse electrodialysis stack [12] and therefore no electrical-
grid power would be needed. Although MECs have been developed for hydrogen and
methane production [13, 14], and microbial fuel cells have been developed to release
orthophosphate from iron phosphate [15], struvite crystallization in an MEC has not yet
been explored. The objectives of this study were therefore to show that struvite could be
precipitated at the cathode of a MEC in the presence of high carbonate concentrations,
and to quantify the affect applied potential and cathode material had on system metrics
such as struvite precipitation, phosphate removal, hydrogen production and efficiency.
3.3 Materials and Methods
3.3.1 Reactor Construction and Operation
Four single chamber MEC reactors (28 mL empty bed volume) were constructed as
previously described [11]. Heat-treated graphite fiber brushes (Gordon Brush, OD = 2.5
cm, L = 2.5) were used as anodes [16, 17]. Stainless steel 304 mesh #60 (SSM, McMaster
Carr, USA) was selected as a cathode based on previous tests showing this mesh size to
be optimum for hydrogen evolution in MECs. Both SSM and a stainless steel flat plates
(SSF, 0.09 mm thick, Trinity Brand Industries, IL, USA) were tested to determine the
relative effects of surface area on hydrogen production rates, struvite crystal precipitation
and phosphate removal. An area of 7 cm2 was used for both the SSM and SSF cathodes in
all calculations based on projected area. However, the SSF cathodes had an exposed
48
surface area of 19 cm2 due to the mesh fibers, and thus the SSM had a surface area to
projected area ratio of As/Ap = 2.7 [18]. Cathodes were placed 2-cm from the anode at the
opposite side of the chamber (Figure 3.1).
The reactors were inoculated with effluent from working microbial fuel cells. Two
solutions were used: an ammonia phosphate solution (5.96 g/L NaHCO3 1 g/L
NaC2H3O2, 0.46 g/L NH4H2PO4, trace vitamins and minerals) used during startup; and a
magnesium chloride solution (5.96 g/L NaHCO3, 1 g/L NaC2H3O2, 1.66 g/L MgCl2, trace
vitamins and minerals) used in struvite precipitation tests. The concentrations of
carbonate (75 mM) and phosphate (4 mM) were selected to mimic those commonly
found in anaerobic digestion supernatant at domestic wastewater treatment plants [19].
During start-up, all reactors were fed the ammonia phosphate solution and operated in
fed-batch mode at 30°C and an applied potential of 0.9 V until voltages and gas
production were reproducible over three batch cycles. Once stable, two of the four
reactors were converted to operate as microbial electrochemical struvite cells (MESCs)
by feeding a solution containing 4 mM concentrations of the struvite ionic components
(Mg2+, NH4+, and PO4
3-). The struvite reactor feed solution was mixed before each feed
cycle to prevent crystallization during storage, and was composed by combining equal
aliquots of the ammonia phosphate and magnesium chloride solutions. The four reactors
(2 control, 2 MESC) were run at applied voltages (Eap) of 0.75, 0.90, and 1.05 V for at
least three batch cycles with either SSM or SSF cathodes. The four reactors were also
operated at the open circuit voltage (OCV) for three cycles (each 24 hours) to determine
if struvite precipitated at the cathode with no current.
49
3.3.2 Analytical Techniques
Voltage was added to the reactor circuits with DC power supplies (Circuit Specialists,
Inc., USA). Current was calculated using Ohm’s law by measuring the voltage drop
across a resistor (10 Ω) using a multimeter (Keithley Instruments, OH) [11]. Cathode
potentials were determined using a reference electrode (Ag/AgCl, +0.197 V). Gas
production, gas composition [11], hydrogen production (Q, m3/m3-d), coulombic
efficiency (CE), electrical efficiency (ηe), and overall system efficiency (ηe+s) were
calculated as previously described [20], where the CE is based on the difference in the
initial and final solution chemical oxygen demand (COD). The electrical energy input per
kg of COD removed (Whin, kWh/kg-COD) was determined for each condition:
𝑊ℎ!" =!!"!"
!!!!∆!"#∙!
(3.1)
where Win is the previously described [20] electrical input and ∆COD is the change in
solution chemical oxygen demand and V is the volume of solution fed to the reactor. The
energy recovered as hydrogen per kg of COD (WhH2, kWh/kg-COD) was determined
with equation 3.2:
𝑊ℎ!! = 𝑌!! ∙ 𝐻𝐻𝑉!! (3.2)
where YH2 (kg H2/kg-COD) is the is the hydrogen yield [21], HHVH2 is the higher heating
value of hydrogen (39.4 kWh/kg-H2) [22].
Crystal accumulation (g/m2-hr) on the cathode surface was measured gravimetrically
(before and after each cycle) using a high precision balance (Mettler Toledo AG104,
USA), with the mass normalized by the fed-batch cycle time and the cathode projected
surface area (7 cm2). Used cathodes were dried at room temperature overnight before
50
being weighed. Crystals were imaged and characterized using scanning electron
microscopy coupled with energy dispersive spectroscopy (Hitachi S-3500N SEM EDS
BSE, Japan). Crystals were verified to be struvite by comparing energy dispersive
spectrographs (EDS) of a pure struvite standard (99.998%, Alfa Aesar) to spectra
collected from cathode accumulated crystals. Phosphate removal due to struvite
precipitation was determined for filtered (0.2 µm, Restek, USA) samples obtained before
and after each batch cycle, using ion chromatography (Dionex-120, AS-16 anion
exchange column, CA).
Electrochemical impedance spectroscopy (EIS, Biologic Science Intruments, VMP 3,
France) was used to investigate the effects of struvite precipitation on cathode kinetics.
An abiotic three electrode cell (working electrode (WE), cathode; reference electrode
(REF), Ag/AgCl; counter electrode, Pt ring) filled with 75 mM carbonate electrolyte was
used in all EIS experiments. To replicate MEC experimental conditions, the WE
potential was set at the average operating potential (Figure S2) for each cathode (SSM
and SSF) at each applied voltage (0.75, 0.90, and 1.05 V).
3.4 Results and Discussion
3.4.1 Cathode Crystal Accumulation and Phosphate Removal
Phosphate removal was achieved by struvite crystallization in MESCs. In
comparison to reactors operated in open circuit (OCV), phosphate removal was
significantly higher in cells where hydrogen evolution and crystal precipitation occurred
51
(Figure 3.2). Removal rates increased with applied voltages for both SSM and SSF
cathodes. However, the rate of increase in phosphate removal was much higher for
MESCs with mesh than flat plate cathodes. At Eap = 0.75 V phosphate removal in MESC
cells operated with SSM (18 ± 10%) and SSF (20 ± 12%) cathodes were very similar, but
as the applied voltage increased to 0.90 V and 1.05 V, SSM cathode reactors out-
performed SSF cathode reactors. At 1.05 V, phosphate removal reached 38 ± 9% with
SSM cathode reactors at 1.05 V, in comparison to 26 ± 13% in SSF cathode reactors.
Struvite crystals precipitated on both types of cathodes at all applied cell voltages.
Crystal accumulation rates on SSM cathodes increased linearly with applied voltage from
0.32 ± 0.05 g/m2-hr at 0.75 V, to 0.85 ± 0.09 g/m2-hr at 1.05 V (Figure 3.2). The rate of
crystal accumulation on SSF cathodes increased with applied voltage (from 0.25 ± 0.05
g/m2-hr at 0.75 V, to 0.53 ± 0.09 g/m2-hr at 1.05 V) at significantly lower rates than
observed with SSM cathodes. The difference in accumulation rates was most likely due
to the significantly higher surface area of mesh than flat cathodes. Cathodes did not
increase in mass when MESC reactors were operated in OCV, or in control MECs fed
magnesium deficient solution.
3.4.2 Crystal Analysis
Crystals that accumulated on the cathodes were analyzed by SEM-EDS to examine
the morphology of the crystal as well as elemental composition. Crystal growth did not
result in a uniform scale layer but rather produced blooms growing away from the
cathode surface (Figure 3.3). On the SSM cathodes, crystal blooms were concentrated at
wire junctions in the woven mesh (Figure 3.3a) compared to more sporadic growth on the
52
SSF cathodes (Figure 3b). Crystals on all cathodes displayed the needle shaped prismatic
morphology typical of struvite [23]. Energy dispersive spectra of crystals on SSM
(Figure 3d) and SSF (Figure 3.3e) cathodes (at all applied voltages) verified the
composition as that of struvite based on comparison with pure struvite standards (Figure
3.3f).
3.4.3 MEC Performance
Hydrogen production (Figure 3.4a) in MESCs with SSM cathodes ranged from Q =
1.0 ± 0.3 m3-H2/m3-d (0.75 V) to 2.3 ± 0.2 m3-H2/m3-d (1.05 V). Hydrogen produced by
MESCs with SSF cathodes increased from 0.7 ± 0.02 m3-H2/m3-d (0.75 V) to 2.0 ± 0.6
m3-H2/m3-d (1.05 V). Hydrogen production rates for both MESCs and control reactors
with SSM and SSF cathodes increased linearly with applied voltage. Production values
for control reactors (fed magnesium deficient solution) with SSM and SSF cathodes were
slightly lower but very similar to reactors with struvite production, with differences
ranging from 3% (1.05 V) to 16% (0.75 V). This indicates that struvite crystal
precipitation in MESCs did not negatively affect hydrogen production rates.
The most significant difference in hydrogen production was observed between
reactors with SSM and SSF cathodes. Reactors with SSM cathodes produced hydrogen at
a greater rate than SSF cathode reactors at all applied voltages because reactors with SSM
cathodes operated at higher densities than reactors with SSF cathodes. The difference in
production decreased from 50% at 0.75 V to 11% at 1.05 V implying that the SSF
cathodes could be limited by surface area at lower applied voltages.
53
Current density (A/m2-cathode, Figure 3.4b) increased linearly with applied voltage
for all reactors. Reactors with mesh cathodes produced slightly more current than those
with sheet reactors. Struvite precipitating reactors produced more current that control
MECs. This difference is most likely attributed to differences is anode performance
between reactors.
Coulombic efficiencies (CEs) were very near or above 100% for all tested conditions
(Figure 5a). The CE in control MECs decreased linearly with applied voltage from 125 ±
14% at 0.75 V to 114 ± 5% at 1.05 V. Generally, struvite reactors had lower CEs than
control reactors and did not show a linear relationship to applied voltage. MESC reactors
with sheet cathodes had an average CE of 111 ± 7%. MESCs with mesh cathodes had an
average CE of 101 ± 8%. Coulombic efficiencies above 100% indicate electron cycling
via hydrogen gas (current generation due to hydrogen oxidation by anodic bacteria)
occurred within both MEC and MESC reactors. Electron cycling results in excess current
generation and additional input of external energy which negatively effects electrical
efficiency [24].
The electrical efficiency of hydrogen production in MESC (the ratio of energy
recovered as hydrogen to the electrical energy consumed by the power supply) exceeded
100% at all applied voltages (Figure 3.5b). The electrical efficiency of control MEC
reactors with SSM and SSF cathodes decreased linearly with each increase in applied
voltage. Observed values of electrical efficiency were similar (within 10%) for MEC and
MESC reactors with both SSM and SSF cathodes at 0.90 V (165% – 181%) and 1.05 V
(139% – 161%). Even with high CE values, the electrical efficiencies presented in this
study suggest that the energy demands of struvite production with an MEC could be
54
significantly offset by energy recovered as hydrogen. This could substantially lower the
production cost of struvite precipitation in wastewater treatment systems.
The overall system efficiency (ηe+s), which compares the energy recovered as
hydrogen gas to the energy input as electricity and substrate, ranged from 67 – 78% for
MESCs and 78 – 83% for control MECs (Figure 3.5c). MESC reactors with both mesh
and sheet cathodes peaked in overall efficiency at 0.90 V. At this applied voltage interval,
the overall hydrogen production efficiency of struvite precipitating and control reactors
with both mesh and sheet cathodes were within 6% of each other. MESC reactors with
mesh cathodes were only 4 – 6% less efficient than control reactors at all applied
voltages. This similarity indicates that system efficiency in reactors with mesh cathodes
were not significantly affected by struvite crystallization. There was a significant
disparity in efficiency between control MECs and MESCs with sheet cathodes at 0.75 V
(19%) and 1.05 V (15%). The reason for this difference was that MESCs with sheet
cathodes had lower hydrogen yields than MECs with sheet cathodes indicating that
struvite crystallization negatively affected catalysis at the surface of sheet cathodes.
Energy production normalized to COD removal was higher than electrical
consumption at all tested conditions (Figure 6a-c). Electrical energy input rate increased
with applied voltage and ranged from 3.1 ± 0.3 kWh/kg-COD at 0.75 V to 4.6 ± 0.3
kWh/kg-COD at 1.05 V. Electrical energy consumption values were higher for control
reactors that MESC reactors. The observed rates of energy input in this study are higher
than the estimated for activated sludge treatment of organic wastewater (~1 kWh/kg-
COD) [25]. Electricity consumption was amplified by electron cycling within the reactor
as well as the high applied voltages required to overcome the electrochemical over-
55
potential of the stainless steel cathodes. Energy production in the form of hydrogen
(which is related to acetate concentration) did not follow an obvious trend and ranged
from 4.2 ± 0.3 kWh/kg-COD at 0.75 V to 5.8 ± 0.3 kWh/kg-COD at 1.05 V. Energy
recoveries were lower in MESC reactors than control MECs. The most significant
difference was observed at 0.75 V where MESC reactors with SSM cathodes were 17%
lower and MESCs with SSF cathodes were 29% lower than control MECs. The net
energy recovered as hydrogen for control MEC reactors decreased as applied voltage
(electricity input) increased ranging from 3.1 ± 0.3 kWh/kg-COD at 0.75 V to 4.6 ± 0.3
kWh/kg-COD at 1.05 V. Energy recovery in MESCs reactors with both SSM and SSF
cathodes peaked at 0.9 V. This peak in energy recovery implies that the presence of
crystals on the MESC cathodes effected hydrogen recovery at 0.75 and 1.05 V.
3.4.4 Electrochemical Impedance Analysis of Cathodes
Used mesh and flat plate cathodes from both MEC and MESCs were analyzed with
electrochemical impedance spectroscopy to quantify the effect of crystal accumulation
based on changes in individual factors that contribute to overall internal resistance.
Charge transfer resistances (Rct) were very similar for all cathodes at all potentials, at 1.7
± 0.4 Ω (range of 1.5 – 2.8 Ω) (Figure A3a-c). As expected, there was little change in
solution resistance (Rs= 21 ± 1.4 Ω). As cathode potential decreased, large changes in
diffusion resistance were observed for the SSM cathodes, with decreases from Rd = 41 ±
1.3 Ω at a cathode potential of –0.95 V (0.75 V) to 21 ± 2.8 Ω at –1.2 V (1.05V).
Changes in the diffusion resistance were similar for SSM cathodes with struvite
precipitation (MESC) compared to the control cathodes without struvite crystals (MEC).
56
In contrast, as cathode potential decreased there was little change in the diffusion
resistance using both the struvite precipitation (MESC) and control (MEC) SSF cathodes.
The similarities of internal resistance between MEC and MESC reactor cathodes provides
evidence that although energy recovery was affected by crystal precipitation, it did not
affect hydrogen evolution kinetics.
3.5 Conclusions
Phosphorus was successfully precipitated as struvite (0.3 – 0.9 g/m2-hr) in single
chamber MESCs with up to 40% of soluble phosphate removed at high electrical energy
efficiencies (135% – 181%). The hydrogen production rates (0.7 – 2.3 m3-H2/m3-d) and
electrical energy efficiencies obtained in this study suggest the energy demands of
struvite production with an MESC would be significantly offset by energy recovered as
hydrogen. This could substantially lower the operational costs for struvite recovery.
57
3.6 Figures
Figure 3.1: Single chamber MESC reactor schematic.
Figure 3.2: Struvite Crystal Growth and phosphate removal in MESC with stainless steel mesh (SSM) and shim (SSF) cathodes.
58
c)
d) e) f)
a) b)
2
59
Figure 3.3: (a) Hydrogen production rate (Q) and (b) current density of single chamber MEC (C) and MESC (S) reactors.
b)
a)
60
Figure 3.4: (a) Coulombic Efficiency (CE), (b) electrical efficiency (ηe) and (c) overall efficiency (ηe+s) of single chamber MEC (C) and MESC (S) reactors.
a)
b)
c)
61
Figure 3.5: Energy comsumption and recovery for SSM and SSF in MEC (C) and MESC (S) reactors a) 0.75 b) 0.90 and c) 1.05 V.
a)
b)
c)
62
3.7 Literature Cited
1. Gilbert, N., The disappearing nutrient. Nature, 2009. 461(8): p. 716-718.
2. Cordell, D., J.O. Drangert, and S. White, The story of phosphorus: Global food security
and food for thought. Global Environmental Change, 2009. 19(2): p. 292-305.
3. Plaza, C., et al., Greenhouse Evaluation of Struvite and Sludges from Municipal
Wastewater Treatment Works as Phosphorus Sources for Plants. Journal of Agricultural
and Food Chemistry, 2007. 55(20): p. 8206-8212.
4. Massey, M.S., et al., Effectiveness of recovered magnesium phosphates as fertilizers in
neutral and slightly alkaline soils. Agronomy Journal, 2009. 101(2): p. 323-329.
5. Ohlinger, K.N., T.M. Young, and E.D. Schroeder, Predicting struvite formation in
digestion. Water Research, 1998. 32(12): p. 3607-3614.
6. Doyle, J.D. and S.A. Parsons, Struvite formation, control and recovery. Water Research,
2002. 36(16): p. 3925-3940.
7. Suzuki, K., et al., Removal of phosphate, magnesium and calcium from swine wastewater
through crystallization enhanced by aeration. Water Research, 2002. 36(12): p. 2991-
2998.
8. Le Corre, K.S., et al., Struvite crystallisation and recovery using a stainless steel
structure as a seed material. Water Research, 2007. 41(11): p. 2449-2456.
9. Jaffer, Y., et al., Potential phosphorus recovery by struvite formation. Water Research,
2002. 36(7): p. 1834-1842.
10. Moussa, S.B., et al., Electrochemical precipitation of struvite. Electrochemical and Solid-
State Letters, 2006. 9: p. C97.
11. Call, D. and B.E. Logan, Hydrogen production in a single chamber microbial electrolysis
cell (MEC) lacking a membrane. Environmental Science & Technology, 2008. 42(9): p.
3401-3406.
63
12. Kim, Y. and B.E. Logan, Hydrogen production from inexhaustible supplies of fresh and
salt water using microbial reverse-electrodialysis electrolysis cells. Proceedings of the
National Academy of Sciences, 2011.
13. Cheng, S., et al., Direct biological conversion of electrons into methane by
electromethanogenesis. Environmental Science & Technology, 2009. 43(10): p. 3953-
3958.
14. Clauwaert, P. and W. Verstraete, Methanogenesis in membraneless microbial electrolysis
cells. Applied Microbiology and Biotechnology, 2009. 82(5): p. 829-836.
15. Fischer, F., et al., Microbial fuel cell enables phosphate recovery from digested sewage
sludge as struvite. Bioresource Technology, 2011. 102(10): p. 5824-5830.
16. Wang, X., et al., Use of Carbon Mesh Anodes and the Effect of Different Pretreatment
Methods on Power Production in Microbial Fuel Cells. Environmental Science &
Technology, 2009. 43(17): p. 6870-6874.
17. Logan, B.E., et al., Graphite fiber brush anodes for increased power production in air-
cathode microbial fuel cells. Environmental Science & Technology, 2007. 41(9): p.
3341-3346.
18. Zhang, Y., M.D. Merrill, and B.E. Logan, The use and optimization of stainless steel
mesh cathodes in microbial electrolysis cells. International Journal of Hydrogen Energy,
2010. 35(21): p. 12020-12028.
19. Battistoni, P., et al., Phosphorus removal from a real anaerobic supernatant by struvite
crystallization. Water Research, 2001. 35(9): p. 2167-2178.
20. Logan, B.E., et al., Microbial electrolysis cells for high yield hydrogen gas production
from organic matter. Environmental Science & Technology, 2008. 42(23): p. 8630-8640.
64
21. Cusick, R.D., P.D. Kiely, and B.E. Logan, A monetary comparison of energy recovered
from microbial fuel cells and microbial electrolysis cells fed winery or domestic
wastewaters. International Journal of Hydrogen Energy, 2010. 35(17): p. 8855-8861.
22. NAE, N.a., The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D
Needs2004, Washington, D. C.: The National Academies Press.
23. Le Corre, K.S., et al., Impact of calcium on struvite crystal size, shape and purity. Journal
of Crystal Growth, 2005. 283(3-4): p. 514-522.
24. Call, D.F., R.C. Wagner, and B.E. Logan, Hydrogen production by Geobacter species
and a mixed consortium in a microbial electrolysis cell. Applied and Environmental
Microbiology, 2009. 75(24): p. 7579-7587.
25. Rabaey, K. and W. Verstraete, Microbial fuel cells: novel biotechnology for energy
generation. Trend Biotechnol, 2005. 23(6): p. 291-298.
Chapter 4
Electrochemical Struvite Precipitation from Digester Effluent with a
Fluidized Bed Cathode Microbial Electrolysis Cell
4.1 Abstract
Microbial electrolysis cells (MECs) can be used to simultaneously convert wastewater
organics to hydrogen and precipitate struvite, but scale formation at the cathode surface
can block catalytic active sites and limit extended operation. To promote bulk phase
struvite precipitation and minimize cathode scaling, a two-chamber MEC was designed
with a fluidized bed to provide particles with a high surface area for crystal growth and to
enable particle cleaning of the cathode surface. Current generation with this design
elevated cathode pH between 8.3 – 8.7 under continuous flow conditions. Soluble
phosphorus removal using digester effluent ranged from 70 – 85%, compared to 10 –
20% for the control (open circuit conditions). At low current densities (≤ 2 mA/m2),
scouring of the cathode by fluidized particles prevented scale accumulation over a period
of 8 days. There was nearly identical removal of soluble phosphorus and magnesium
from solution, and an equimolar composition in the collected solids, supporting
phosphorus removal by struvite formation. At an applied voltage of 1.0 V, energy
consumption from the power supply and pumping (0.2 Wh/L, 7.5 Wh/g-P) was
significantly less than that needed by other struvite formation methods based on pH
adjustment such as aeration and NaOH addition. Total energy consumption could be
66
completely offset if hydrogen recoveries >80% could be achieved. In the anode chamber,
current generation led to COD oxidation (1.1 – 2.1 g-COD/L-d) and ammonium removal
(7 – 12 mM) from digestate amended with 1 g/L of sodium acetate. These results indicate
that a fluidized bed cathode MEC is a promising method of sustainable electrochemical
nutrient and energy recovery method for nutrient rich wastewaters.
4.2 Introduction
Wastewater treatment plants use enhanced biological phosphorus removal (EBPR) to
minimize nutrient discharge into waterways. Stabilization of excess activated and EBPR
sludge in an anaerobic digester results in significant releases of nutrients, minerals, and
alkalinity (Ca2+, Mg2+, K+, PO43-, NH4
+, and HCO3–) [1]. During digestion, 40 − 80% of
the phosphorus contained in EBPR sludge is solubilized, leading to nutrient recycling
within the plant and calcium and magnesium phosphate salt precipitation within the
digester and on equipment surfaces [2, 3].
In nutrient rich digester effluent, a number of sparingly soluble salts can form (Table
4.1). The most prevalent are magnesium ammonium phosphate (struvite), calcium
carbonate (calcite) and amorphous calcium phosphate (ACP) [4]. Precipitation occurs
when the ionic product of a salt exceeds the solubility limit (aka super saturation) [5].
Since the ionic product of phosphate salts are pH dependent, precipitation takes place
where carbonate striping raises solution pH, such as pumps and conduits due to the high
turbulence [6]. Scale formation on pumps and within conduit can lead to reduced pipe
capacity, conduit replacement, and costly equipment failure [7-9]. The annual operational
cost of dealing with scale formation at wastewater treatment plants has been estimated at
67
roughly $10,000 per million gallon per day (MGD) of dry weather capacity [10]. The
potential economic benefit of preventing scale formation while recovering renewable
sources of nitrogen and phosphorus has led to the development of side-stream phosphate
salt crystallization systems [6, 8, 11]. The majority of systems have focused on struvite
crystallization because both nitrogen and phosphorus are removed.
Fluidized bed reactors (FBRs) are most commonly used for crystallized phosphate
recovery because the design creates an abundance of reactive surface area and solution
turbulence [12]. These conditions promote rapid molecular growth at the crystal surface
interface which eliminates the induction time associated with nucleating crystal directly
from solution [13]. Struvite precipitation within crystallization reactors affected by pH
and molar concentration ratios. The minimum solubility of struvite has been reported at a
pH of 10.7, but effective secondary crystal growth has been observed at pH values greater
than 8.0 [9]. When solution pH within an FBR is increased to 8.5 or greater, soluble
phosphate removal often reaches 70 − 90% of at low hydraulic retention times (1 − 4
hours) [11, 12, 14, 15]. Struvite precipitation is adversely affected by the presence of
calcium and carbonate. When solution Ca:P ratios are equal to or greater than Mg:P
ratios, struvite crystal purity, size and phosphorus removal can be compromised by
amorphous calcium phosphate co-precipitation [16-18]. In digester effluent, high
alkalinity can lead to magnesium carbonate complexation, which reduces the amount of
magnesium available for struvite precipitation [19, 20].
Despite high nutrient removal efficiencies, proliferation of phosphate salt
crystallization systems has been limited by the monetary cost of increasing solution pH.
Sodium hydroxide (NaOH) is often used because it is highly soluble and can rapid
68
increase the solution pH [7, 11, 16, 21]. The costs of chemical addition can drive
phosphorus recovery costs (~$3,500/ton-P) to be significantly higher than the value of
nutrients contained in struvite (~$765/ton-P)[9, 22]. Stripping carbonate alkalinity by
aeration has also been shown to effectively raise pH and reduce cationic complexation
with carbonate [17, 23], but blowers are energy intensive and aeration releases ammonia
into the atmosphere.
Recently, microbial electrochemical technologies (METs), such as microbial fuel
cells (MFCs) and electrolysis cells (MECs) have been explored as a chemical free and
energy efficient method of precipitation struvite [24-26]. METs consist of an anode,
where anaerobic microbes oxidize organic matter and transfer electrons to an external
circuit, and a cathode where the electrons and protons catalytically combine by reducing
oxygen to water (MFC) or producing hydrogen gas (MEC). Proton consumption
increases pH at the cathode surface and induces struvite precipitation [27].
Simultaneously removing organic matter and nutrients from wastewater while generating
renewable energy and fertilizer is in concept a sustainable approach for struvite
production, but initial results with batch-fed, single-chamber reactors have shown low
phosphorus removal efficiencies (20 – 50%) and cathode failure due to scale
accumulation.
In this study, a two-chamber MEC reactor with a fluidized bed cathode chamber was
developed to enhance phosphorus removal and minimize cathode scaling. To induce a
bulk phase pH increase in the cathode chamber, a cation exchange membrane separated
the electrodes. The cathode chamber was operated as a fluidized bed to increase struvite
precipitation through production of high surface area particles. In addition, particles
69
scoured the cathode and helped to minimize scale accumulation. To investigate the effect
of current density on struvite precipitation rate and cathode scaling, the FBR-MEC was
operated under continuous flow at three applied cell voltages (0.8V, 1.0V, and 1.4V). At
each applied voltage, the influent flow rate was adjusted to maintain a cathode pH near
8.5. The reactor was also operated at open circuit (control) to evaluate phosphorus
removal within the fluidized bed in the absence of current.
4.3 Materials and Methods
4.3.1 Reactor Construction
The MEC was composed of two concentric cylindrical chambers. A tubular cation
exchange membrane separated the inner cathode chamber from the outer anode chamber.
The inner cathode chamber was designed to replicate the dimensions of fluidized bed
columns previously used for struvite precipitation. The fluidization section of the cathode
had a diameter to cross sectional area ratio of 25 (inner diameter of d=1.1 cm, height of h
= 28 cm) [12, 28]. The fluidization chamber consisted of a cylindrical cation exchange
membrane (d = 1.1 cm, h = 15 cm) and sections of clear PVC tubing at each end of the
membrane tube (d = 1.1, h = 15 cm). The cation exchange membrane tube was
constructed by applying a CEM monomer solution to the surface of a high-density
polyethylene filtration membrane (Porex, USA) with 30 µm diameter pores. The CEM
layer created an ionic connection between chambers but prevented convective liquid
flow. Above the fluidization section, an expansion vessel (d = 3.8 cm, h = 7.9 cm) was
constructed to reduce liquid velocity and promote solids retention [23]. To collect biogas
produced at the cathode surface, a cut glass anaerobic tube was glued to a section of the
70
1.1 cm diameter PVC above the expansion section and liquid exit port [29]. The total
cathode volume was 160 mL. The cathode electrode was a 3 cm × 15 cm piece of
stainless steel 304 mesh (30x30, McMaster Carr, US) rolled into a 1 cm diameter cylinder
and placed within the fluidization column. The outer anode chamber was contained a
carbon mesh anode [30], and had an empty bed volume of 180 mL.
4.3.2 Solutions
Anaerobic digester effluent samples were collected from the top section of the
secondary digester at a local domestic wastewater treatment plant (University Park, PA)
at least once per week and stored at 4°C. The treatment plant is comprised of two
secondary treatment trains (activated sludge and trickling filters) followed by biological
nutrient removal. The cathode chamber was fed raw digestate. The organic matter in the
digestate at the point of collection was not adequate for MEC current generation. To
provide sufficient electron donor to the anode microbial community, 1 g/L of sodium
acetate was added to the digestate fed to the anode chamber and the COD increased from
0.6 g/L to 1.3 g/L (Table 2).
4.3.3 Reactor Operation
Solutions were pumped from reservoirs to the MEC reactor through Viton tubing
(Masterflex L/S size 16, USA) using variable speed peristaltic pumps (Cole Parmer,
USA). Solution was recycled in both the anode and cathode chambers at a flow rate of 20
mL/min. Recirculation in the cathode chamber created an up-flow velocity of 0.3 cm/s,
71
which was sufficient to fluidize a bed of glass particles (30 – 50 µm diameter,
Polyscience, Inc, USA).
The MEC was operated at applied cell voltages (Eap) of 0.8, 1.0 and 1.4 V using an
external power supply (BK Precision, USA). The reactor was also operated under open
circuit (control) to establish a baseline for ion removal in the absence of electrolytic pH
changes. The digestate influent flow rate for each applied voltage was adjusted to
maintain a cathode pH near 8.5. The influent flow rates for both the anode and cathode
chambers were: 0.4 mL/min (0.8 V), 0.7 mL/min (1.0 V), and 1.0 mL/min (1.4 V). At
each applied cell voltage condition, the MEC was continuously operated for 8 days.
4.3.4 Solution and Electrochemical Measurements
To monitor current generation, the voltage drop across a 10 Ω resistor placed within
the MEC circuit was measured every 15 minutes using a multimeter (Keithley
Instruments, USA) connected to a computer. Catholyte pH was measured every 15
minutes by multi-parameter probe placed in a flow cell within the recirculation flow path.
The pH probe was calibrated before each run. Liquid samples (50 mL) were collected
from the influent reservoir as well as anode and cathode effluent ports each day.
Phosphorus, magnesium, calcium, potassium, and sodium concentrations were
determined by inductively coupled plasma atomic emission spectroscopy (ICP-AES)
(Perkin Elmer Optima 5300 DV, USA). To determine total ion concentrations, samples
(10 mL) were acidified to dissolve any fine particles prior to filtration (0.45 µm pore
diameter). Soluble ion concentrations were acidified after filtration to prevent
precipitation during sample storage. Ammonium concentrations were determined with an
72
ammonia probe (Thermo Scientific Orion, USA). Carbonate alkalinity was determined by
titration to pH 4.5 with 20 mM H2SO4 according to standard methods [31]. Chemical
oxygen demand was determined with a chromic acid colorimetric method (Hach, CO).
4.3.5 Analysis
Phosphorus conversion efficiency (X) accounts for both fine phosphate salt particles
leaving the reactor as well as solids retained and was calculated as the difference between
reactor influent total P and effluent soluble P as:
𝑋 = 100 !"!"!!"!"#!"!"
(4.1)
where tPin is total influent phosphorus (M), and sPout is the effluent soluble phosphorus
concentration (M). Precipitation efficiency (L), which measures the extent of precipitate
retained within the fluidized bed, was calculated as the difference between influent and
effluent total P concentrations as:
𝐿 = 100 !"!"!!"!"#!"!"
(4.2)
The collection efficiency (η), which is used to determine the retention of converted
particles within the fluidized bed, is the ratio of conversion to precipitation (X/L) [17, 23].
The rate of conversion and precipitation were calculated for each sample interval using:
𝑚! = (!"!"!!"!"#)!!"#
!!"#∙∆! (4.3)
𝑚! = (!"!"!!"!"#)!!"#
!!"!∙∆! (4.4)
where mx is the phosphorus mass conversion rate (g-P/m3cat-d), mL is the
phosphorus precipitation rate (g-P/m3cat-d), Vdig is the volume of digestate treated by the
73
cathode chamber during the sample interval (m3), Vcat is the cathode volume (m3), and ∆t
is the sample interval (d).
Energy input (Ein) to the MEC, from pumping and applied potential was calculated
for each sample interval (Win = WCell + Wpump). Energy input based on current generation
and applied cell potential (Wcell) was calculated as previously described [29]. Energy
input for the recirculation and influent feed lines were calculated as previously described
based on hydraulic head and flow rates [32]. At the end of each run, solids were collected
from the FBR cathode. A portion of the collected solids from each run was dissolved in
an acidic solution (0.2 M HCl) to determine elemental composition. Solid dissolution
supernatant was filtered and analyzed using ICP-AES. Super saturation of sparingly
soluble salts and ionic complexation within the disgestate were predicted at pH values
ranging from 7 – 11 using an electrolyte modeling software package (OLI, USA). The
same software package was also used to estimate the NaOH dose required to raise the
raw digestate to a pH of 8.5.
4.4 Results and Discussion
4.4.1 Phosphorus Removal in the Fluidized Bed Cathode
Enhanced phosphorus removal and elevated pH were maintained in the fluidized bed
cathode MEC. By generating current from soluble organic matter, solution pH within the
cathode chamber was maintained between 8.3 and 8.7 under continuous flow conditions
(Figure 4.3a). Under elevated pH conditions, phosphorus conversion (X: 70 – 85%) and
precipitation efficiency (L: 66 – 71%) within the FBR cathode chamber were
significantly higher than open circuit conditions (Figure 4.2b). Without current
74
generation cathode solution pH (7.3 ± 0.1) did not increase as it passed through the
fluidized bed and phosphorus conversion (8 ± 6%) and precipitation within the reactor (8
± 6%) were negligible (Figure 4.2b). The level phosphorus removal achieved by the FBR
MEC was comparable to previous crystallization reactor investigations in which solution
pH was increased by aeration or chemical base addition [8, 11, 23] and is a marked
improvement over previous MET struvite investigations.
4.4.2 Phosphorus Precipitation in the Fluidized Bed Cathode
The rate of solids precipitation within the FBR was dependent on influent flow rate
and current density. Current density increased with applied voltage (Figure 4.3a), which
allowed the cathode influent flow rate to be raised while maintaining elevated solution
pH and phosphorus removal within the FBR. To maintain elevated pH at the lowest
applied potential and current density (0.8 V: 0.8 ± 0.03 A/m2-membrane) solution flow
rate was set to 0.4 mL/min (θh = 11.1 hrs), resulting in a precipitation rate of 67 ± 3
grams of phosphorus per cubic meter of cathode volume per day (Figure 4.4a). At 1.0 V,
the current density (1.7 ± 0.2 A/m2) and flow rate (Qf,c = 0.7 mL/min, θh = 4.8 hrs) were
approximately twice that produced at the lower applied voltage, and the phosphorus
precipitation rate increased to 146 ± 15 g-P/m3cat-day. At the applied voltage (1.4 V, 2.4 ±
0.4 A/m2-membrane) the flow rate was 1.0 mL/min (θh = 2.8 h), producing an average
phosphorus precipitation rate of 210 ± 14 g-P/m3cat-day. The highest rate is within the
range previously obtained in fluidized bed crystallization reactors with chemical base
addition to increase pH [8]. Under open circuit conditions, the phosphate precipitation
rate was 16 ± 15 g-P/m3cat-day at an influent flow rate of 0.7 mL/min.
75
Solids analysis indicated that struvite was the dominant precipitate in the MEC
cathode. Acidic dissolution analysis of solids recovered from the FBR cathode indicated
that phosphorus to magnesium that were equal to, or slightly greater than 1 for all of the
elevated pH conditions (Table 4.4). Molar Mg:P ratios greater than one at applied
voltages of 1.0 and 1.4 V suggest that Newberyite may have co-precipitated struvite. At
OCV, the Mg:P ratio of recovered solids was 0.8 suggesting solids accumulated near
neutral pH were primarily magnesium phosphate salts. Calcium, potassium and sodium
were present in all acid dissolution samples, but at very low concentrations.
Struvite precipitation may have been promoted by favorable influent ionic
concentration ratios. The Mg:P ratio in digestate fed to the reactor varied slightly (1.3 –
1.5) and were above a common target ratio (1.2) which has led to efficient struvite
precipitation in previous investigations [33]. Ratios of Mg:Ca and Ca:P greater than 2
have previously been shown promote small diameter (<10 um) amorphous calcium
phosphate precipitation [17, 34] which easily leave fluidization bed reactors and decrease
precipitation efficiency. Relatively low influent magnesium to calcium (1.1 – 1.2) and
phosphorus to calcium ratios (1.2 – 1.4) also likely contributed to high collection
efficiency.
4.4.3 Cathode Scaling
Cathode performance was maintained by particle scouring at current densities less
than 2 A/m2-membrane (Figure 3). At 0.8 and 1.0 V, current density and cathode pH
were held stable through the entire continuous flow run. As seem in SEM images, the
active surface area SS mesh cathodes operated at 0.8 and 1.0 V was maintained after 8
76
days of operation (Figure 5 b and c). Minor scale accumulation was localized to the wire
junctions. At 1.4 V, the initial current density of 2.8 ± 0.2 A/m2 began to decrease after 3
days of operation and reduced to 2 A/m2 by the end of the 8-day run. By the end of the
1.4 V run, scale had completely covered the cathode surface. It is unclear whether scale
accumulation was caused by localized pH increase at higher current density or by higher
precipitation rates. It is possible that larger particles and higher up-flow velocity could
maintain the surface area at higher precipitation rates.
4.4.4 Molar Ionic Removal in the Fluidized bed Cathode
Soluble magnesium and phosphorus removal from digestion effluent fed to the
cathode were nearly equimolar, providing further evidence that struvite precipitation was
the dominant P removal pathway within the fluidization bed. The similarity in molar
phosphorus (0.9 – 1.1 mM) and magnesium (0.9 – 1.0 mM) removal for each applied
potentials interval also implied that hydraulic retention time, which decreased from 11
hours at 0.8 V to 2.8 hours at 1.4 V, did not influence the extent of struvite precipitation
within the cathode chamber. When magnesium flux from the anode chamber is
considered, the molar Mg:P removal ratios for all of the elevated pH conditions are all
near unity (Table 4.3). Co-precipitation of calcium phosphates did not occur in the MEC
cathode and soluble calcium concentrations in cathode solution actually increased due to
transport across the CEM during MEC current generation (Figure 4.2a). Soluble
phosphorus concentration in anode chamber solution did not increase, indicating that P
removal in the cathode was caused by precipitation and that the CEM maintained high
selectivity through out the study (Figure 4.6a).
77
Sodium acetate amendment to anode digestate produced a concentration gradient
between anode and cathode, which led to increasing sodium flux across the CEM with
HRT (Figure 4.2a). Measurement of ammonium flux into the cathode solution was highly
variable due to volatilization caused by elevated pH and turbulence. Ammonium
transport from anode to cathode was measured with greater accuracy from changes in
anode concentration. Ammonia removal from the anode increased with current density
and flow rate, ranging from 7.0 ± 1.3 mM at 0.8 V to 12 ± 2.3 mM at 1.4 V (Figure 4.6b).
4.4.5 Electrolyte Modeling
The influent digestate was modeled using electrolyte software to gain a better
understanding of why magnesium phosphate, rather than calcium phosphates, formed in
the fluidized bed. Modeling showed at pH ≥ 8, both magnesium and calcium phosphates
became supersaturated. At the operating pH of the FBR cathode, several salts were
supersaturated including calcite, hydroxyapatite (HAP), struvite and amorphous calcium
phosphate (ACP). Although hydroxyapatite was the most saturated of the sparingly
soluble phosphate salt (Figure 4.9b), crystalline HAP does not form directly. The
precursor, ACP, transforms into a homogeneous crystalline structure over a period of
days, which is unlikely to occur within the hydraulic retention time of the FBR [19, 35].
Despite high saturation of HAP, the competition for soluble phosphorus results primarily
in the formation of ACP and struvite. At the operating pH of the FBR cathode, model
results shows that struvite was much more saturated than ACP. Additionally, ACP and
calcite precipitation rates can be severely inhibited by CO32-, Mg2+, soluble organic
ligands and organic particles [19, 36-40].
78
4.4.6 COD Removal in the Anode Chamber
MEC current was generated by microbial oxidation of organic matter at the anode.
COD removal rate in the anode chamber increased with current density ranging from 1.1
± 0.01 g-COD/L-d at 0.8V to 2.1 ± 0.2 g-COD/L-d at 1.4 V (Figure 4.6c). Coulombic
efficiency ranged from 20 – 30%, indicating that acetoclastic methanogenesis likely
contributed to COD removal.
4.4.7 Energy Consumption
Energy consumption during MEC operation resulted almost entirely from the power
used to set an applied potential, and as affected by current density, precipitation rate, and
influent flow rate. At 0.8 V, the energy consumption normalized to the amount of
phosphorus precipitated was 6.5 ± 0.5 Wh/g-P (0.20 ± 0.4 Wh/L). When the applied
potential was increased to 1.0 V, a substantial increase in both current and precipitation
rate offset the increase energy needed at the higher applied voltage, and therefore the
normalized energy input (6.6 ± 0.4 Wh/g-P, 0.22 ± 0.01 Wh/L) was similar to that
obtained at 0.8 V. At 1.4 V, the increase in current density and precipitation rate were
insufficient to compensate for the increased energy, resulting in a much higher energy
consumption of 10 ± 1.0 Wh/g-P (0.3 ± 0.07 Wh/L).
Recovery of hydrogen produced in the MEC could offset electrical energy demand.
Based on the current produced. Hydrogen produced at the MEC cathode could not be
properly collected during continuous flow experiments. It is possible that produced
hydrogen diffused through the PVC reactor body and fitting. If cathodic hydrogen
79
recovery from the FBR-MEC could reach 80%, energy in the hydrogen gas produced
(based on the lower heating value) would exceed energy consumption (Figure 4.7). Even
without hydrogen gas recovery, energy consumption by the FBR-MEC remain below that
needed for struvite recovery using other pH adjustment methods. Energy consumption
rates observed here were 85 – 90% lower than a system in which aeration was used to
increase digestate pH to 8.4 (~65 Wh/g-P, 1 Wh/L) [4]. It is estimate, based on
electrolyte modeling of the cathode influent digestate, that 5 mM (200 mg/L) of NaOH
would be required to achieve an equivalent pH increase to 8.5 (Figure 4.9a). Considering
that the embodied electrical energy of caustic soda produced by the Chlo-Alkali industry
(114 Wh/mol-NaOH, [41]), equivalent pH and phosphorus removal (1.0 V case) by
caustic addition would represent 0.57 Wh/L and 17.5 Wh/g-P of energy consumption.
The energy required for operation of the FBR-MEC could be further improved in
several ways (Figure 4.8a). For example, constructing the cathode chamber with a
thinner CEM would significantly reduced ionic transport resistance which has limited
MEC hydrogen production in previous investigations [42]. Selecting a catalyst with less
activation over-potential such as nickel foam [43] or molybdenum disulfide [44] would
also reduce applied potential and energy consumption. With the system used in this study
energy consumption normalized to phosphorus precipitation would be significantly
reduced by treating waste with higher influent P concentrations (Figure 4.8b).
4.5 Conclusions
The results of this study indicate that continuous microbial electrochemical struvite
precipitation from digester supernatant is possible if the cathode is operated as a fluidized
80
bed and the current if the scouring rate is sufficient to prevent cathode scaling. Mg:P
molar ratios of soluble ionic removal and recover solids indicated that struvite formation
was the primary method of phosphorus removal within the FBR cathode. MEC energy
consumption (0.2 - 0.3 Wh/L) was lower than conventional pH adjustment systems used
to recover struvite, such as aeration and chemical base addition. Reducing internal
resistance and producing hydrogen with a low over-potential catalyst could optimize
energy consumption. Capturing MEC produced biogas at a cathodic efficiency of 80%
could offset energy consumption.
81
4.6 Tables
Table 4.1: Common name, formula and solubility constant of sparing soluble salts with potential to form in digester effluent.
Table 4.2: Solution characteristics of digestate fed to cathode and anode chambers.
Parameter Cathode Anode
pH 7.2 ± 0.1 7.2 ± 0.1
COD (g/L) 0.61 ± 0.1 1.3 ± 0.1
tC (mg CaCO3/L) 2100 ± 120 2300 ± 130
NH4-N (mg/L) 490 ± 45 510 ± 52
tP (mg/L) 46 ± 3 46 ± 3
sP (mg/L) 45 ± 3 45 ± 3
sMg (mg/L) 51 ± 3 52 ± 4
sCa (mg/L) 76 ± 4 76 ± 3
sK (mg/L) 57 ± 4 60 ± 5
sNa (mg/L) 230 ± 15 476 ± 30
Mg:P 1.4 ± 0.1 1.5 ± 0.1
Ca:P 1.3 ± 0.1 1.3 ± 0.1
Mg:Ca 1.1 ± 0.03 1.1 ± 0.1
Common Name Formula pKso Reference
Struvite MgNH4PO4·6H2O(s) 13.2 [6]
Newberyite MgHPO4·H2O(s) 18.2 [45]
Bobierrite Mg3(PO4)2·22H2O(s) 25.2 [45]
Calcium Hydrogen phosphate CaHPO4(s) 6.7 [5]
β-Amorphous Calcium Phosphate (ACP) β-Ca3(PO4)2(s) 24.0 [5]
Hydroxyapatite (HAP) Ca5(PO4)3OH(s) 55.9 [5]
Calcite CaCO3(s) 8.3 [5]
82
Table 4.3: Cathode effluent ionic concentrations.
FBR Cathode Effluent Ionic Removal Ratios Applied Voltage
tP (mg/L)
sP (mg/L)
sMg (mg/L)
sCa (mg/L) Mg:P Ca:P
OCV 46 ± 3 46 ± 2 49 ± 2 76 ± 2 1.2 0.6
0.8V 14 ± 1 13 ± 1 34 ± 1 87 ± 2 1.0 0.0
1.0V 13 ± 3 7.9 ± 2 23 ± 2 71 ± 3 0.99 0.0
1.4V 15 ± 1 13 ± 3 31 ± 2 76 ± 3 0.98 0.0
Table 4.4: Molar composition of collected solids as determined by acid dissolution.
Solid Sample Mg:P Ca:P K:P Na:P Glass Beads 0 0 0 0
OCV 0.84 0.00 0.08 0.01 0.8 V 1.00 0.01 0.04 0.02 1.0 V 1.06 0.05 0.01 0.01 1.4 V 1.06 0.07 0.02 0.03
83
4.7 Figures
Figure 4.1: Process flow of fluidized bed cathode microbial electrolysis cell with close up of electrode interface.
Raw digestate
Anod
e FB
R C
atho
de
Recycle Pump
Cathode Effluent Anode
Effluent
Digestate w/ 1 g/L NaAc
Car
bon
Mes
h An
ode
Tubu
lar C
atio
n Ex
chan
ge M
embr
ane
SS M
esh
Cat
hode
Car
bon
Mes
h An
ode
Tubu
lar C
atio
n Ex
chan
ge M
embr
ane
SS M
esh
Cat
hode
Fluidized Particle Bed
CO2
Power Supply
COD Biofilm
e- e-
84
Figure 4.2: a) Molar ionic removal and b) soluble (sP) and total (tP) phosphorus removal in the fluidized bed cathode.
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
Ion
Rem
oval
in C
atho
de (m
M) P Mg Ca K Na
0
20
40
60
80
100
OCV 0.8V 1.0V 1.4V
Phos
phor
us R
emov
al (%
) sPtP
b)
a)
85
0.0
1.0
2.0
3.0
4.0
0 1 2 3 4 5 6 7 8
Cur
rent
Den
sity
(A/m
2 )
Days
0.8V 1.0V 1.4V
7.0
7.5
8.0
8.5
9.0
0 1 2 3 4 5 6 7 8
Cat
hode
pH
Days
0.8V 1.0V1.4V OCV
0
10
20
30
40
50
60
0 1 2 3 4 5 6 7 8
Cat
hode
Effl
uent
sP
(mg/
L)
Days
OCV 0.8V 1.0V 1.4V
Figure 4.3: a) MEC current density (b) cathode pH and (c) effluent soluble P concentrations during eight day continuous flow experiments.
a) b)
c)
86
Figure 4.4: a) Phosphorus precipitation and collection rate in the fluidized bed cathode and (b) energy input normalized to precipitated and collected phosphorus.
0
50
100
150
200
250
OCV 0.8V 1.0V 1.4V
Rem
oval
Rat
e (g
-P/m
3 -rx
tr-d) Precipitation Rate
Collection Rate
0
2
4
6
8
10
12
14
OCV 0.8V 1.0V 1.4V
Ener
gy In
put (
Wh/
g-P)
PrecipitatedCollected
a)
b)
87
Figure 4.5: SEM images of stainless steel mesh cathodes operated at a) open circuit, b) 0.8 V, c) 1.0 V and d) 1.4 V.
a) b)
c) d)
88
Figure 4.6: a) Ionic, (b) ammonium and (c) COD removal in the anode chamber.
0.0
0.5
1.0
1.5
2.0
2.5
OCV 0.8V 1.0V 1.4V
CO
D R
emov
al (g
-CO
D/L
-rxtr-
d)
-10
-5
0
5
10
15
20
NH
4+R
emov
al fr
om A
node
(mM
)-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
Ion
Rem
oval
in A
node
(mM
)P Mg Ca K Na
a)
b)
c)
89
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0 20 40 60 80 100
Ener
gy B
alan
ce (W
h/L)
Cathodic Recovery (%)
H2Input
Figure 4.7: Modeled energy balance and a function of cathodic hydrogen recovery.
90
0123456789
10
0 50 100 150 200
Influent Conc. (mg-P/L)
0
2
4
6
8
10
0.0 0.2 0.4 0.6 0.8 1.0
Ene
rgy
Inpu
t (kW
h/kg
-P)
Applied Voltage (V)
Figure 4.8: Modeled energy input predictions as a function of a) applied voltage and b) influent phosphorus concentration.
a) b)
91
Figure 4.9: Electrolyte modeling of a) equivalent NaOH addition and b) super-saturation of sparingly soluble salts, and c) ionization in the digestate.
0
10
20
30
40
50
6 7 8 9 10 11E
quiv
alen
t NaO
H A
dditi
on (m
M)
pH
-3
-2
-1
0
1
2
3
6 7 8 9 10 11
log
(IP/K
so)
pH
CaCO3
MgNH4PO4
β-‐Ca3(PO4)2(s)
0.0
0.2
0.4
0.6
0.8
1.0
7 8 9 10 11
Ioni
zatio
n Fr
actio
n
pH
Ca2+
CaH2PO4+
Mg2+
MgCO3(o)
NH4+
PO43-‐
CaCO3(o)
HPO42-‐
H2PO4-‐
a)
b)
c)
92
4.8 Literature Cited
1. Pérez-Elvira, S., P.N. Diez, and F. Fdz-Polanco, Sludge minimisation technologies.
Reviews in Environmental Science and Bio/Technology, 2006. 5(4): p. 375-398.
2. Jardin, N. and H. Popel, Phosphate release of sludges from enhanced biological P-
removal during digestion. Water science and technology, 1994. 30(6): p. 281-292.
3. Liao, P., D.S. Mavinic, and F.A. Koch, Release of phosphorus from biological nutrient
removal sludges: A study of sludge pretreatment methods to optimize phosphorus release
for subsequent recovery purposes. Journal of Environmental Engineering and Science,
2003. 2(5): p. 369-381.
4. Battistoni, P., et al., P removal from anaerobic supernatants by struvite crystallization:
long term validation and process modelling. Water Research, 2002. 36(8): p. 1927-1938.
5. Snoeyink, V.L. and D. Jenkins, Water chemistry1980: John Wiley.
6. Ohlinger, K.N., T.M. Young, and E.D. Schroeder, Predicting struvite formation in
digestion. Water Research, 1998. 32(12): p. 3607-3614.
7. Ohlinger, K.N., T.M. Young, and E.D. Schroeder, Kinetics effects on preferential struvite
accumulation in wastewater. Journal of Environmental Engineering, 1999. 125: p. 730.
8. Ohlinger, K.N., T.M. Young, and E.D. Schroeder, Postdigestion struvite precipitation
using a fluidized bed reactor. Journal of Environmental Engineering, 2000. 126: p. 361.
9. Doyle, J.D. and S.A. Parsons, Struvite formation, control and recovery. Water Research,
2002. 36(16): p. 3925-3940.
10. Neethling, J. and M. Benisch, Struvite control through process and facility design as well
as operation strategy. Water science and technology: a journal of the International
Association on Water Pollution Research, 2004. 49(2): p. 191.
93
11. Bhuiyan, M., D. Mavinic, and F. Koch, Phosphorus recovery from wastewater through
struvite formation in fluidized bed reactors: a sustainable approach. Water Science &
Technology, 2008. 57(2): p. 175-181.
12. Seckler, M.M., O.S.L. Bruinsma, and G.M. Van Rosmalen, Phosphate removal in a
fluidized bed--I. Identification of physical processes. Water Research, 1996. 30(7): p.
1585-1588.
13. Bhuiyan, M.I.H., D. Mavinic, and R. Beckie, Nucleation and growth kinetics of struvite
in a fluidized bed reactor. Journal of Crystal Growth, 2008. 310(6): p. 1187-1194.
14. Seckler, M.M., O.S.L. Bruinsma, and G.M. Van Rosmalen, Calcium phosphate
precipitation in a fluidized bed in relation to process conditions: A black box approach.
Water Research, 1996. 30(7): p. 1677-1685.
15. Forrest, A., et al., Optimizing struvite production for phosphate recovery in WWTP.
Journal of Environmental Engineering, 2008. 134: p. 395.
16. Le Corre, K.S., et al., Impact of calcium on struvite crystal size, shape and purity. Journal
of Crystal Growth, 2005. 283(3-4): p. 514-522.
17. Moerman, W., et al., Phosphate removal in agro-industry: Pilot- and full-scale
operational considerations of struvite crystallization. Water Research, 2009. 43(7): p.
1887-1892.
18. Hao, X.D., et al., Struvite formation, analytical methods and effects of pH and Ca(2 + ).
Water Sci Technol, 2009. 59(6): p. 1077-1083.
19. Cao, X. and W. Harris, Carbonate and Magnesium Interactive Effect on Calcium
Phosphate Precipitation. Environmental Science & Technology, 2007. 42(2): p. 436-442.
20. Mañas, A., et al., Parameters influencing calcium phosphate precipitation in granular
sludge sequencing batch reactor. Chemical Engineering Science, 2012. 77(0): p. 165-
175.
94
21. Le Corre, K.S., et al., Struvite crystallisation and recovery using a stainless steel
structure as a seed material. Water Research, 2007. 41(11): p. 2449-2456.
22. Dockhorn, T. About the economy of phosphorus recovery. in International conference on
nutrient recovery from wastewater streams. 2009.
23. Battistoni, P., et al., Phosphorus removal from a real anaerobic supernatant by struvite
crystallization. Water Research, 2001. 35(9): p. 2167-2178.
24. Cusick, R.D. and B.E. Logan, Phosphate recovery as struvite within a single chamber
microbial electrolysis cell. Bioresource Technology, 2012. 107(0): p. 110-115.
25. Hirooka, K. and O. Ichihashi, Phosphorus recovery from artificial wastewater by
microbial fuel cell and its effect on power generation. Bioresource Technology, 2013.
137(0): p. 368-375.
26. Ichihashi, O. and K. Hirooka, Removal and recovery of phosphorus as struvite from
swine wastewater using microbial fuel cell. Bioresource Technology, 2012. 114(0): p.
303-307.
27. Moussa, S.B., et al., Electrochemical precipitation of struvite. Electrochemical and Solid-
State Letters, 2006. 9: p. C97.
28. Seckler, M., et al., Phosphate removal in a fluidized bed--II. Process optimization. Water
Research, 1996. 30(7): p. 1589-1596.
29. Call, D. and B.E. Logan, Hydrogen production in a single chamber microbial electrolysis
cell (MEC) lacking a membrane. Environmental Science & Technology, 2008. 42(9): p.
3401-3406.
30. Wang, X., et al., Use of Carbon Mesh Anodes and the Effect of Different Pretreatment
Methods on Power Production in Microbial Fuel Cells. Environmental Science &
Technology, 2009. 43(17): p. 6870-6874.
95
31. APHA, ed. Standard Methods for the Examination of Water and Wastewater. 20th ed.,
ed. L.S. Clesceri, A.E. Greenberg, and A.D. Eaton1998, American Public Health
Association, American Water Works Association, Water Environment Federation:
Washington DC.
32. Zhang, F., et al., Long-Term Performance of Liter-Scale Microbial Fuel Cells Treating
Primary Effluent Installed in a Municipal Wastewater Treatment Facility. Environmental
Science & Technology, 2013. 47(9): p. 4941-4948.
33. Nelson, N.O., R.L. Mikkelsen, and D.L. Hesterberg, Struvite precipitation in anaerobic
swine lagoon liquid: effect of pH and Mg: P ratio and determination of rate constant.
Bioresource Technology, 2003. 89(3): p. 229-236.
34. Le Corre, K.S., et al., Impact of calcium on struvite crystal size, shape and purity. Journal
of Crystal Growth, 2005. 283(3): p. 514-522.
35. Abbona, F. and A. Baronnet, A XRD and TEM study on the transformation of amorphous
calcium phosphate in the presence of magnesium. Journal of Crystal Growth, 1996.
165(1–2): p. 98-105.
36. Ferguson, J.F., D. Jenkins, and J. Eastman, Calcium Phosphate Precipitation at Slightly
Alkaline pH Values. Journal (Water Pollution Control Federation), 1973. 45(4): p. 620-
631.
37. Ferguson, J.F. and P.L. McCarty, Effects of carbonate and magnesium on calcium
phosphate precipitation. Environmental Science & Technology, 1971. 5(6): p. 534-540.
38. Van der Houwen, J.A., et al., The effect of organic ligands on the crystallinity of calcium
phosphate. Journal of Crystal Growth, 2003. 249(3): p. 572-583.
39. Reddy, M.M., Crystallization of calcium carbonate in the presence of trace
concentrations of phosphorus-containing anions: I. Inhibition by phosphate and
96
glycerophosphate ions at pH 8.8 and 25 C. Journal of Crystal Growth, 1977. 41(2): p.
287-295.
40. Reddy, M.M. and K.K. Wang, Crystallization of calcium carbonate in the presence of
metal ions: I. Inhibition by magnesium ion at pH 8.8 and 25 C. Journal of Crystal
Growth, 1980. 50(2): p. 470-480.
41. Pellegrino, J.L., Energy and environmental profile of the US chemical industry2000:
Office of Industrial Technologies, Energy Efficiency and Renewable Energy, US
Department of Energy.
42. Rozendal, R.A., et al., Effect of the type of ion exchange membrane on performance, ion
transport, and pH in biocatalyzed electrolysis of wastewater. Water Science and
Technology, 2008. 57(11): p. 1757-1762.
43. Jeremiasse, A.W., et al., Ni foam cathode enables high volumetric H< sub> 2</sub>
production in a microbial electrolysis cell. International Journal of Hydrogen Energy,
2010. 35(23): p. 12716-12723.
44. Tokash, J.C. and B.E. Logan, Electrochemical evaluation of molybdenum disulfide as a
catalyst for hydrogen evolution in microbial electrolysis cells. International Journal of
Hydrogen Energy, 2011. 36(16): p. 9439-9445.
45. Dempsey, B.A. Removal and reuse of ammonia and phosphate by precipitation of
struvite. in PROCEEDINGS OF THE INDUSTRIAL WASTE CONFERENCE-PURDUE
UNIVERSITY. 1998. MOSBY YEAR BOOK EUROPE LTD.
Chapter 5
Energy Capture from Thermolytic Salt Solutions in Microbial Reverse
Electrodialysis Cells
5.1 Abstract
A system was developed capable of converting wastewater organics and low-grade
thermal energy (such as solar, geothermal, and waste heat) to electricity by using a
thermolytic salt, ammonium bicarbonate, as the saline solutions in a microbial reverse
electrodialysis cell (MRC). To examine the potential of this system, ammonium
bicarbonate solutions of a high concentration (HC) and low concentration (LC) were used
to generate electricity in a lab-scale MRC. The reverse-electrodialysis stack containing 5
pairs of HC and LC cells (HC = 1.1 M; LC = 0.011 M) directly contributed 2.4 ± 0.01
W/m2 (43%) to the maximum power, and also enhanced power recoveries from electrode
reactions. The electrode power density (from acetate oxidation) increased from 1.08 ±
0.03 W/m2 in a single chamber microbial fuel cell (MFC) to 3.3 ± 0.04 W/m2 in the
MRC. Maximum power density of the MRC reached 5.6 W/m2, which was 20% higher
than that reported with artificial seawater and river water. When fed domestic
wastewater, the MRC peak power density was 3.0 ± 0.05 W/m2, with the MFC electrodes
contributing 2.0 ± 0.05 W/m2, the highest reported value of electricity production from
domestic wastewater. The results of this study show the potential of using ammonium
98
bicarbonate salt solutions to recovery energy from low grade thermal and wastewater
with a MRC.
5.2 Introduction
Worldwide energy demand is expected to increase to 23 TW by 2050 [1]. The
development of 10 TW of renewable energy [1] is needed to sustain a habitable
ecosystem while meeting the energy demand of a human population projected to eclipse
9 billion by mid-century [2]. Organic wastewater and salinity energy are two sources of
renewable energy that remain essentially untapped. Microbial fuel cells (MFCs) have
recently emerged as an efficient method of directly producing electricity from organic
matter present in wastewater (the treatment of which currently accounts for 1% of total
energy use in the U.S.[3])[4, 5]. Reverse electrodialysis (RED) is a salinity-gradient
energy system designed to convert the entropic energy released by mixing solutions of
different ionic-strength into electrical energy within a fuel cell arrangement [6-8]. A
recently proven combination of these technologies, the microbial reverse electrodialysis
cell (MRC), resulted in a syntrophic enhancement of energy recovery from both
wastewater and salinity-gradient energy [9].
MFCs capture electrical current produced by exoelectrogenic microbes that oxidize
soluble organic matter present in wastewater [10, 11]. MFCs contain anodes where
microbes grow and release electrons, and air cathodes, where electrons are consumed in
the oxygen reduction reaction [4]. Spontaneous power production in single chamber air
cathode MFCs (normalized to projected cathode area) has reached 2.7 W/m2 under
idealized conditions [12] but is substantially lower when reactors are fed complex and
99
poorly conductive substrates such as domestic wastewater (0.26 – 0.45 W/m2) [13-15].
When coupled with a RED stack, the MRC produced 4.3 W/m2 when fed buffered acetate
as well as seawater and river water; however, its potential applications are limited only in
estuary areas or treatment plants that discharge treated effluent into the ocean [9].
In a RED system, a stack of alternating anion- and cation-exchange membranes
separating solutions of high concentration (HC) and low concentration (LC) is
constructed between fuel cell electrodes. The additive electrochemical junction potential
(0.1 – 0.2 V per cell pair for typical seawater and fresh water) [16] of the electrolytic-pile
provides a driving force for current generation at the electrodes. Because of this energy
losses at the electrodes, the net output energy of an RED system is highly dependent on
energy conversion at the electrodes especially under high current conditions [17]. By
coupling a RED stack with spontaneous MFC electrode kinetics (i.e., oxidation of
organic matter at the anode and oxygen reduction at the cathode), significant energy
losses which occur with water electrolysis (up to 2.5 V in cell voltage loss in a 25
membrane stack due to electrode overpotentials) [18] and system damage associated with
the hexacyanoferrate redox couple ([Fe(CN)6]3-/[Fe(CN)6]4-) such as membrane
poisoning, iron precipitation, and HCN gas production at the electrodes can be avoided
[17, 19-22]. The spontaneous nature of MRC current generation also allows salinity-
gradient energy generation with much fewer membranes (5 membrane pairs compared to
25 – 50) [9] representing a substantial reduction in the membrane associated capitol costs
($100 – 200/m2) [8].
At the time of this work, RED technology had only been proposed to harness energy
released by the mixing of seawater and fresh water limiting the application to estuaries
100
(1.7 TW) [23], or for the case of MRC technology to ocean side wastewater treatment
plants (18.5 GW) [8]. If RED saline solutions were composed of ammonium bicarbonate,
which decomposes to ammonia and carbon dioxide gas at low temperatures (40 – 60°C),
[24] MRCs can create electricity from abundant sources of low grade thermal energy
such as waste heat (204 GW), [25] solar (120,000 TW) and geothermal (12 TW) energy
[1]. In this study, MRC power production using ammonium bicarbonate saline solutions
is investigated. The effects of saline solution concentration, salinity ratio (HC/LC), and
flow rate on power density were examined with both synthetic wastewater containing
acetate and effluent from primary clarifier at a local domestic wastewater treatment plant.
5.3 Materials and Methods
5.3.1 Reactor Construction
The lab scale MRC reactor was constructed as previously described [9] with minor
modifications. The 4-cm cubic anode chamber (Lexan, 30 mL empty bed volume)
contained a graphite brush anode (D = 2.7 cm, L = 2.3 cm, Mill-Rose Labs Inc., OH).
The brush anode was heat treated [26] before it was inoculated with the effluent from an
existing MFC and enriched in a conventional single chamber MFC prior to MRC
operation. The cathode chamber (2-cm cubic chamber, 18 mL empty bed volume)
contained a Pt loaded (0.5 mg Pt/cm2) carbon cloth passive air cathode constructed as
previously described [27] with a Nafion catalyst binder on the water side and four layers
of polytetrafluoroethylene diffusion layers on the air side. The cathode chamber also
101
served as the first flow channel of the high concentrate salt stream to prevent the pH rise
in the cathode chamber.
The RED stack, assembled between the anode and cathode chambers, consists of with
6 anion- and 5 cation-exchange membranes (Selemion AMV and CMV, Asahi glass,
Japan), creating 5 pairs of alternating HC and LC chambers as previously described[9].
Inter-membrane chambers were sealed and separated by silicon gaskets, each with an 8
cm2 (2 × 4 cm) rectangular cross section cut out. Inter-membrane chamber width (1.3
mm) was maintained with a 2 cm2 (0.5 × 4 cm) strip of polyelthylene mesh. The total
ionic exchange membrane area in the RED stack was 88 cm2. The total MRC empty bed
volume was 58.4 mL (RED stack + Cathode = 28.4 mL; Anode = 30 mL). The HC
solution entered the reactor at the cathode and flowed serially through the 5 HC cells in
the stack, exiting from the cell next to the anode chamber. The LC stream entered the
RED stack near the anode and flowed serially through the 5 LC cells in the stack, exiting
from the cell next to the cathode chamber. A peristaltic pump (Cole Parmer, IL)
continuously fed the HC and LC solutions at a flow rate of 1.6 mL/min, unless specified
otherwise. During power density curve experiments fresh saline solutions were pumped
through the RED stack with the effluent collected in separate reservoirs. In batch recycle
experiments 0.1 L of each solution was recycled through the stack in airtight flow paths.
Before each batch the stack and tubing were flushed with matching solutions.
5.3.2 Solutions
Ammonium bicarbonate HC solution was prepared by dissolving ammonium
bicarbonate salt (Alfa Aesar, MA) into deionized water within an airtight vessel. The
102
initial HC tested was 1.8, 1.1, 0.95, 0.8, and 0.5 M. The LC solutions were prepared
depending on given salinity ratios of 50, 100, and 200 by diluting an aliquot of the HC
solution. The anode solution contained 1 g/L of sodium acetate, in 50 mM carbonate
buffer (4.2 g/L NaHCO3-) containing 0.231 g/L NH4H2PO4 and trace vitamins and
minerals [28]. Domestic wastewater was collected from the primary clarifier of the Penn
State University wastewater treatment plant.
A second order relationship between ammonium bicarbonate solution concentration
and solution conductivity (determined by conducting a stepwise dilution series, Figure
S4) was used to estimate initial and final concentrations of HC and LC streams.
Conductivity and pH of the HC and LC streams were measured (Mettler-Toledo, OH)
before and after each batch recycle experiment.
5.3.3 Power Measurement
Power production in batch recycle system efficiency experiments was determined by
measuring the potential drop across a fixed external resistance (300 Ω) for both MRC and
single chamber MFC operations. Voltage drop was recorded every 20 minutes by a
digital multimeter (Keithley Instruments, OH). Electrical current (i) was determined by
Ohm’s law. Power was calculated by multiplying the electrical current and total cell
voltage. To determine the maximum MRC power (PMRC) production at each condition the
reactor was held at open circuit voltage for one hour and then the external resistance was
decreased from 1,000 to 50 Ω every 20 minutes with the voltage read at each resistance
interval. Power contribution by the electrode reactions (PMFC) was determined by
measuring the anode potential (Ean) and cathode potential (Ecat) against Ag/AgCl
103
reference electrodes (BASi, IN): PMFC = (Ecat – Ean)× i. The RED stack power
contribution was calculated by finding stack voltage (Vstk) with two reference electrodes
located on both ends of the stack as: PRED = Vstk × i.
The MRC anode was transferred to a single chamber MFC to determine baseline
power production in batch mode and peak power production from carbonate buffered
acetate and domestic wastewater.
5.3.4 Analysis
Coulombic efficiency was determined as previously described [27]. Energy recovery
(rE) is defined by the ratio of energy produced by the MRC reactor and the energy input
as substrate and salinity gradient as written in Eq. (3). Energy efficiency (ηE) is the ratio
of energy produced over the energy consumed as substrate and salinity gradient, were
determined for batch recycle experiments [19]:
𝑟! =!!"#$
!!,!∙∆!!!∆!!"#,!∙ 100% (5.1)
𝜂! =!!"#$
(!!,!!!!,!)∙∆!!!(∆!!"#,!!∆!!"#,!)∙ 100% (5.2)
where EMRC is the energy produced per batch (kJ), ns is the moles of substrate (acetate)
fed to the anode initially (0) and at the end of the batch cycle (f), ∆Gs is the Gibb’s free
energy of substrate (acetate = –846.6 kJ/mol [29], domestic wastewater = 17.8 kJ/g-COD
[30]), ∆Gmix is the free energy that can be created by mixing of HC and LC solutions until
the two solutions reach equilibrium concentration as:
∆𝐺!"# = 𝑅𝑇 (𝑉!"𝑐!,!"𝑙𝑛!!,!"#!!,!"
+ 𝑉!"𝑐!,!"𝑙𝑛!!,!"#!!,!"
)! (5.3)
104
where R is the ideal gas constant (8.314 J/mol-K), T is solution temperature, V is the
volume of solution, c is the molar concentration of ionic species i in the solution, and a is
the activity of species i in the solution.
At a neutral pH, concentrated ammonium bicarbonate is dominated by ammonium
(NH4+) and bicarbonate (HCO3
-) ions but significant amounts of carbamate (NH4CO3-)
and carbonate (CO32-) also contribute to ionic strength. Species specific concentrations
and activities were estimated with OLI Stream Analysis software (OLI Systems, Inc.,
Morris Plains, NJ) at a pH of 7 and temperature of 25 °C (Figures S4 and 5).
To determine ammonia transport into the anode, total ammonia nitrogen (TAN, NH3
+ NH4+) concentration in the substrate was determined before and after each fed-batch
cycle (HACH, CO) [31]. Free ammonia concentration (FAN) was determined also
determined for each fed-batch cycle:
𝑁𝐻! = 𝑇𝐴𝑁 1+ !"!!"
!"! !.!"!#$! !"!#.!"
!
!!
(5.4)
5.4 Results and Discussion
5.4.1 Effect of AmB Concentration on MRC Power Production
At a fixed salinity ratio (SR = 100; HC = 1.1 M; LC = 0.011 M) and flow rate (1.6
mL/min), the MRC produced a maximum power density of 5.6 ± 0.04 W/m2 (0.5 ± 0.003
W/m2 of projected membrane area) (Figure 5.2), 20% higher than the power density
achieved with artificial seawater and freshwater [9]. Peak power was reached at a total
cell voltage was 0.77 V and current density of 0.73 mA/cm2. At this salinity ratio and
105
concentration interval, the RED stack contributed 2.4 ± 0.01 W/m2 (43%) and the
electrode reactions contributed 3.3 ± 0.04 W/m2 (57%) to the total MRC power.
As the molar concentration in the AmB HC solution increased from 0.5 to 1.8 M,
RED stack power increased from 1.2 ± 0.3 to 2.6 ± 0.1 W/m2 (Figure 5.2). By
maintaining salinity ratio (SR 100), the effect of ionic concentration within the stack
could be seen in the polarization curves of the MRC and RED stack at the various HC
concentrations (Figure 5.3, 5.4a). Since the salinity ratio was fixed at 100, the MRC open
circuit voltages of all HCs were identical. However, internal resistance within the MRC
(slope of the cell voltage polarization curves, Figure 5.3) decreased as ammonium
bicarbonate concentration increased, from 170 Ω at 0.5 M to 138 Ω at 1.8 M.
The RED stack directly contributed to MRC power and dramatically enhanced power
production from MFC electrodes. Electrode reactions produced nearly 300% more power
in the MRC than in a conventional single chamber MFC. The power associated with
electrode reactions in the MRC was 3.2 ± 0.2 W/m2 as compared to 1.08 ± 0.03 W/m2
(dashed line of Figure 5.2) from the same electrodes in a single chamber MFC. The stack
enhanced electrode power production by maintaining electrode potentials near open
circuit potentials as current density increased. In the single chamber MFC, the electrode
potentials approached each other (anode increased and cathode decreased) and as current
density increased (Figure 5.4b).
At a HC of 1.8 M maximum power decreased to 4.8 ± 0.1 W/m2. Upon reaching a
current density of 0.6 mA/cm2 (Figure 5.4b) the anode potential rapidly increased
(polarized), significantly reducing the MFC contribution to total MRC power. Anode
polarization was also observed at HC = 1.1 M, but at a much higher current density 0.9
106
mA/cm2, and did not occur at lower HCs of 0.5, 0.8, 0.95 M. Anode polarization at
higher HCs caused permanent damage to the anode biofilm and reparative enrichment
was required to restore performance.
5.4.2 Effect of Salinity Ratio and Flow Rate on MRC Power Production
By varying salinity ratio at a fixed HC of 0.95 M (highest achieved power density
without anode polarization) it was determined that SR 100 produced the highest peak
MRC power production. Since MFC electrode performance was considerably stable over
a wide range of SRs from 50 to 200 (3.0 ± 0.01 W/m2) the difference in MRC peak power
density can be attributed to stack performance (Figure 5.5). Increasing salinity ratio from
100 to 200 increased the resistance of the LC chambers and resulted in a 34% drop in
stack power. Lowering SR to 50 reduced LC solution resistance but also lowered the
RED potential, resulting in a 20% decrease in stack power. With HC solution flowing
through both RED stack flow channels (SR 1), stack power dropped to zero but the MFC
electrodes still produced more power (1.7 ± 0.05 W/m2) in comparison to the single
chamber reactor (1.08 ± 0.03 W/m2). It is likely that the observed power increase was due
to the presence of HC solution in the cathode chamber (65.5 mS/cm conductivity).
Lowering the flow rate (Figure A2.7) from 1.6 to 0.85 mL/min (4.9 ± 0.1 W/m2) had
nearly the same effect as lowering the SR from 100 to 50 (4.7 ± 0.1 W/m2).
5.4.3 Energy Production in Batch Recycle Experiments
To maximize energy recovery and efficiency, HC and LC salt solutions (0.1 L each)
were recycled in airtight flow paths for the duration of anode feeding cycles. During
107
recirculation, the salinity gradient decreased, reducing RED stack power contribution to
total MRC power. As a result, total MRC energy production was similar (108 ± 7 J,
Figure 6b) for all tested HC concentration intervals. Although the stack contribution to
total power decreased, the MRC energy recovery was more than 3× higher than the
energy recovery from acetate in a single chamber MFC reactor (32.6 ± 4 J). The salinity-
gradient energy input increased from 98 ± 1 J at 0.5 M (which was less than the MRC
power production at 0.5 M, 101 ± 2 J) to as high as 360 ± 1 J at 1.8 M. Acetate energy
(193 ± 6 J) was the highest input at all HC concentrations except 1.8 M indicating that
energy recovery and efficiency was mostly limited by the rate and efficiency of substrate
oxidation at the anode.
5.4.4 Energy Recovery and Efficiency
Energy recovery and efficiency were significantly higher in the MRC than in the
single chamber MFC. MRC energy recovery ranged from a high of 30 ± 0.5% at 0.5 M,
to 20 ± 0.01% at 1.8 M (Figure 5.6a). There was little variation in MRC energy efficiency
for HCs below 1.8 M (34 ± 0.5%). At 1.8 M, the MRC produced the lowest measured
energy efficiency (25 ± 0.1). In comparison, both energy recovery (14 ± 2%) and
efficiency (16 ± 2%) the single chamber MFC. The coulombic efficiency of acetate
oxidation in the MRC (66 ± 4%) was also markedly higher than the single chamber MFC
(35 ± 4%) because the RED stack prevented oxygen intrusion into the anode chamber of
the MRC. Reduced energy recovery and efficiency with increasing HCs may be
attributed to volatilization of ammonia and carbon dioxide through the air cathode as well
as osmotic water flux from the LC chambers. Also, the volatility of ammonia and
108
carbonate at high HC concentrations caused observable bubble formation within the
stack, which could have limited the active membrane transport area.
5.4.5 Ammonia Transport into the Anode
Ammonia transport into the anode chamber from the RED stack affected anode
performance for HCs > 1 M. As anodic bacteria produce protons from the oxidation of
organic matter, negative bicarbonate and carbamate ions transported across the terminal
anion exchange membrane to balance charge within the anode chamber. Effluent TAN
concentrations in the anode after batch recycle experiments ranged from 263 ± 32 mg/L
at 0.5 M to 590 ± 36 mg/L at 1.8 M. Effluent FAN, which has been linked to extracellular
polysaccharide destruction in activated sludge [32] and power inhibition in single
chamber MFC [33], ranged from 1.5 ± 0.2 mg/L at 0.5 M to 3.3 ± 0.2 mg/L at 1.8 M
(Figure S8). The effluent TAN concentration for HC of 1.8 M exceeded previously
reported thresholds for power inhibition in MFCs (TAN = 500 mg/L) but inhibition was
observed in power density curves, not in batch recycles experiments.
5.4.6 Power Production from Domestic Wastewater
At a HC concentration of 0.95 M, SR of 100, and 1.6 mL/min flow rate, the peak
power density of the MRC fed domestic wastewater was 2.9 ± 0.05 W/m2 (Figure 5.7a).
The MFC electrode contribution to peak power was 2.0 ± 0.05 W/m2, a 740% increase in
power over when the same electrodes operated in a single chamber MFC fed the same
wastewater (0.27 ± 0.05 W/m2). This MFC electrode power is the highest power ever
109
recorded with domestic wastewater and exceeds a recently reported value for a domestic
wastewater fed MFC with carbon nanotube coated electrodes by ~50% [34].
In batch recycle experiment with domestic wastewater, power production was
sustained for only two hours. As MFC electrode power dropped below the stack, the total
output power quickly reduced to zero (Figure 5.8). This inflection point corresponded to
when the anode potential quickly polarized from negative to positive. COD removal in
batch recycle experiments was 58 ± 5%. The energy production rate during batch recycle
was 0.94 kWh/kg-COD.
5.5 Conclusions
To examine the potential of a heat recovery microbial reverse electrodialysis cell
(MRC), concentrated (HC) and dilute (LC) ammonium bicarbonate solutions at various
concentrations were used to generate electricity in a lab-scale reactor. At an ammonium
bicarbonate HC concentration of 1.1 M, maximum power density reached 5.6 W/m2,
which was 20% higher than reported from an artificial sea and river water MRC. When
fed domestic wastewater, the MRC achieved a peak power density of 2.9 ± 0.05 W/m2
with the electrodes reactions contributing 2.0 ± 0.05 W/m2. The RED stack directly
contribute to total MRC power (2.4 ± 0.01 W/m2 with acetate, 0.8 ± 0.01 W/m2 with
domestic wastewater) and symbiotically enhanced the power production of MFC
electrodes fed acetate (330% increase) and domestic wastewater (740% increase).
110
5.6 Figures
Figure 5.1: (a) Main components of the microbial reverse electrodialysis cell (MRC), showing the membrane stack between the electrodes, the reference electrodes and the circuit containing a load (resistor). (b) Example of how the anion- (AEM) and cation- (CEM) exchange membranes are used to selectively drive the flow of positive ions to the right (towards the cathode) and the negatively-charged ions to the left (towards the anode). (c) Expanded view of the membrane stack showing flow path of the high (HC) and low (LC) concentrate solutions of ammonium bicarbonate. (d) Construction of the gaskets used to direct the flow from one LC chamber to the next LC chamber, avoiding the HC chamber through a short flow path through the membrane and gasket.
111
Figure 5.2: Peak power density of MRC as well as the contributions to total power from the RED stack, and MFC electrodes at various HC concentrations. The dashed line represents peak power density of the same electrodes in a single chamber MFC.
0
1
2
3
4
5
6
0.5 1.0 1.5 2.0
Peak
Pow
er D
ensi
ty (W
/m2 )
HC Concentration (M)
MRC Electrodes RED
MFC
112
Figure 5.3: Polarization curves of the MRC at various HC concentrations as well and the polarization curves of the electrodes in a single chamber MFC.
0.0
0.5
1.0
1.5
2.0
0.0 0.2 0.4 0.6 0.8 1.0
Cel
l Vol
tage
(V)
Current Density (mA/cm2)
MFC0.5M0.8M0.95M1.1M1.8M
113
-0.4
-0.2
0.0
0.2
0.4
0.6
0.0 0.2 0.4 0.6 0.8 1.0
Ele
ctro
de P
onte
ntia
l vs.
SH
E (V
)
Current Density (mA/cm2)
1.1M C 1.1M A 1.8M C
1.8M A 0.95M A 0.95M C
SC C SC A
0.0
0.3
0.5
0.8
1.0
RE
D S
tack
Vol
tage
(V) 1.8M
1.1M0.95M
b)
a)
Figure 5.4: a) RED stack potential as well as b) anode (filled symbols) and cathode (open symbols) potentials, measured during polarization curves, in the single chamber MFC (SC) and the three highest HC concentrations.
114
Figure 5.5: Salinity ratio vs. peak power density at HC concentration of 0.95 M and flow rate of 1.6 mL/min. Dashed line represented peak power density of single chamber MFC fed identical substrate.
0
1
2
3
4
5
6
0 50 100 150 200
Peak
Pow
er D
ensi
ty (W
/m2 )
Salinity Ratio
MRC Electrodes RED
MFCSR1
115
0
10
20
30
40
Rec
over
y/Ef
ficie
ncy
(%)
nErE
MFC
0
50
100
150
200
250
300
350
0.5 1.0 1.5 2.0
Ene
rgy
per B
atch
(J)
HC concentration (M)
Salinity InAcetate InMRC Out
MFC
Figure 5.6: a) Energy recovery, efficiency and b) and energy balance of the MRC vs. HC concentration.
a)
b)
116
-0.4-0.3-0.2-0.10.00.10.20.30.4
0.0 0.2 0.4 0.6
Elec
trode
Pon
tent
ial v
s. S
HE
(V)
Current Density (mA/cm2)
MRC WW A MFC WW AMRC WW C MFC WW C
0.0
1.0
2.0
3.0
Pow
er D
ensi
ty (W
/m2 )
MRC WW
MFC WW
Figure 5.7: a) Peak power density and (b) anode and cathode potentials of MRC and single chamber MFC fed domestic wastewater.
a)
b)
117
Figure 5.8: Batch recycle component (MFC, RED and total MRC) power profile of MRC fed domestic wastewater.
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Pow
er (W
/m2 )
Time (days)
MRCREDMFC
118
5.7 Literature Cited
1. Kamat, P.V., Meeting the clean energy demand: Nanostructure architectures for solar
energy conversion. The Journal of Physical Chemistry C, 2007. 111(7): p. 2834-2860.
2. Division, U.N.P., World Population Prospects: the 2010 Revision, 2010, United Nations.
3. IPCC, Climate Change 2007: Synthesis Report. , in Contribution of Working Groups I, II
and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate
Change2007, IPCC: Geneva.
4. Liu, H., R. Ramnarayanan, and B.E. Logan, Production of electricity during wastewater
treatment using a single chamber microbial fuel cell. Environmental Science &
Technology, 2004. 38(7): p. 2281-2285.
5. Cusick, R.D., P.D. Kiely, and B.E. Logan, A monetary comparison of energy recovered
from microbial fuel cells and microbial electrolysis cells fed winery or domestic
wastewaters. International Journal of Hydrogen Energy, 2010. 35(17): p. 8855-8861.
6. Wick, G.L., Power from salinity gradients. Energy, 1978. 3(1): p. 95-100.
7. Post, J.W., et al., Salinity-gradient power: Evaluation of pressure-retarded osmosis and
reverse electrodialysis. Journal of Membrane Science, 2007. 288(1-2): p. 218-230.
8. Ramon, G.Z., B.J. Feinberg, and E.M.V. Hoek, Membrane-based production of salinity-
gradient power. Energy & Environmental Science, 2011.
9. Kim, Y. and B.E. Logan, Microbial Reverse Electrodialysis Cells for Synergistically
Enhanced Power Production. Environmental Science & Technology, 2011. 45(13): p.
5834-5839.
10. Logan, B.E., Exoelectrogenic bacteria that power microbial fuel cells. Nat Rev
Microbiol, 2009. 7(5): p. 375-81.
11. Logan, B.E., et al., Microbial fuel cells: methodology and technology. Environ Sci
Technol, 2006. 40(17): p. 5181-92.
119
12. Fan, Y., H. Hu, and H. Liu, Sustainable Power Generation in Microbial Fuel Cells Using
Bicarbonate Buffer and Proton Transfer Mechanisms. Environmental Science &
Technology, 2007. 41(23): p. 8154-8158.
13. Ahn, Y. and B.E. Logan, Effectiveness of domestic wastewater treatment using microbial
fuel cells at ambient and mesophilic temperatures. Bioresource Technology, 2010.
101(2): p. 469-475.
14. Liu, H. and B.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, 2004. 38(14): p. 4040-4046.
15. Cheng, S., H. Liu, and B.E. Logan, Increased power generation in a continuous flow
MFC with advective flow through the porous anode and reduced electrode spacing.
Environmental Science & Technology, 2006. 40: p. 2426-2432.
16. Weinstein, J.N. and F.B. Leitz, Electric power from differences in salinity: the dialytic
battery. Science, 1976. 191(4227): p. 557.
17. Veerman, J., et al., Reverse electrodialysis: Performance of a stack with 50 cells on the
mixing of sea and river water. Journal of Membrane Science, 2009. 327(1-2): p. 136-144.
18. Veerman, J., et al., Reverse electrodialysis: Comparison of six commercial membrane
pairs on the thermodynamic efficiency and power density. Journal of Membrane Science,
2009. 343(1-2): p. 7-15.
19. Post, J.W., H.V.M. Hamelers, and C.J.N. Buisman, Energy Recovery from Controlled
Mixing Salt and Fresh Water with a Reverse Electrodialysis System. Environmental
Science & Technology, 2008. 42(15): p. 5785-5790.
20. Vermaas, D.A., M. Saakes, and K. Nijmeijer, Doubled power density from salinity
gradients at reduced inter-membrane distance. Environmental Science & Technology,
2011. 45: p. 7089-7095.
120
21. Vermaas, D.A., M. Saakes, and K. Nijmeijer, Power generation using profiled
membranes in reverse electrodialysis. Journal of Membrane Science, 2011.
22. Lacey, R., Energy by reverse electrodialysis. Ocean engineering, 1980. 7: p. 1-47.
23. Kuleszo, J., et al., The potential of blue energy for reducing emissions of CO2 and non-
CO2 greenhouse gases. Journal of Integrative Environmental Sciences, 2010. 7(S1): p.
89-96.
24. McGinnis, R.L., J.R. McCutcheon, and M. Elimelech, A novel ammonia-carbon dioxide
osmotic heat engine for power generation. Journal of Membrane Science, 2007. 305: p.
13-19.
25. EIA, U.S., Annual Energy Review 2010, DOE, Editor 2010.
26. Wang, X., et al., Use of carbon mesh anodes and the effect of different pretreatment
methods on power production in microbial fuel cells. Environmental Science &
Technology, 2009. 43(17): p. 6870-6874.
27. Cheng, S., H. Liu, and B.E. Logan, Increased performance of single-chamber microbial
fuel cells using an improved cathode structure. Electrochemistry Communications, 2006.
8(3): p. 489-494.
28. Ambler, J., Perfomance of stainless steel 304 cathodes and bicarbonate buffer in a
Microbial Electrolysis Cell using a new methode of gas characterization, in Civil and
Environmental Engineering2010, The Pennsylvania State University: University Park.
29. Rozendal, R.A., et al., Towards practical implementation of bioelectrochemical
wastewater treatment. Trends in Biotechnology, 2008. 26(8): p. 450-459.
30. Heidrich, E.S., T.P. Curtis, and J. Dolfing, Determination of the Internal Chemical
Energy of Wastewater. Environmental Science & Technology, 2010. 45(2): p. 827-832.
31. APHA, ed. Standard Methods for the Examination of Water and Wastewater. 20th ed.,
ed. L.S. Clesceri, A.E. Greenberg, and A.D. Eaton1998, American Public Health
121
Association, American Water Works Association, Water Environment Federation:
Washington DC.
32. Yang, S.-F., J.-H. Tay, and Y. Liu, Inhibition of free ammonia to the formation of aerobic
granules. Biochemical Engineering Journal, 2004. 17(1): p. 41-48.
33. Nam, J.Y., H.W. Kim, and H.S. Shin, Ammonia inhibition of electricity generation in
single-chambered microbial fuel cells. Journal of Power Sources, 2010. 195(19): p. 6428-
6433.
34. Xie, X., et al., Carbon nanotube-coated macroporous sponge for microbial fuel cell
electrodes. Energy & Environmental Science, 2012.
Chapter 6
Minimal RED Cell Pairs Markedly Improve Electrode Kinetics and
Power Production in Microbial Reverse Electrodialysis Cells
6.1 Abstract
Power production from microbial reverse electrodialysis cell (MRC) electrodes is
substantially improved compared to microbial fuel cells (MFCs) by using ammonium
bicarbonate (AmB) solutions in a multiple-membrane stack and the cathode chamber, but
reducing the number of membranes pairs could help to reduce capital costs. We show
here that using only a single RED cell pair (CP), created by operating the cathode in
concentrated AmB, dramatically increased power production normalized to cathode area
from both acetate (Acetate: from 0.9 to 3.1 W/m2-cat) and wastewater (WW: 0.3 to 1.7
W/m2), by reducing solution and kinetic resistances at the cathode. The addition of a
second RED cell pair further increased power production (Acetate: 4.2 W/m2; WW: 1.9
W/m2) and anode biofilm activity. These power densities are close to those previously
achieved using 11 membranes, indicating near optimal electrode performance with only
one or two cell pairs. When normalized to total membrane area, the 1-CP (Acetate: 3.1
W/m2-mem; WW: 1.7 W/m2) and 2-CP (Acetate: 1.3 W/m2; WW: 0.6 W/m2) power
densities were much higher than previous MRCs (0.3 − 0.5 W/m2-membrane with
acetate). The rate of wastewater COD removal, normalized to reactor volume, was 30 –
123
50 times higher in 1-CP and 2-CP MRCs than that in a single chamber MFC. It is
estimated that further reductions in RED solution and transport resistance could increase
MRC power production to 6 – 8 W/m2-cat (2 – 3 W/m2-mem). These findings show that
even a single cell pair AmB RED stack can significantly enhance the electrical power
production and wastewater treatment.
6.2 Introduction
Microbial reverse electrodialysis cells (MRCs) are biotechnologies designed to generate
renewable energy from unconventional sources of organic wastewater and salinity gradients. In
the United States alone, ~155 GWh (1.23 kWh/m3) could be generated from organics in
wastewater [1-3]. Ammonium bicarbonate (AmB) is a thermolytic salt that decomposes to
ammonia and carbon dioxide gas at low temperatures (40 – 60°C) [4]. The low decomposition
temperature of AmB could enable generation of salinity gradients from abundant sources of low
grade thermal energy such as waste heat (204 GW) [5], geothermal (12 TW), and solar (120,000
TW) energy [6].
Within a MRC, the same electrodes used in a microbial fuel cell (MFC) are placed on each
side of a reverse electrodialysis (RED) membrane stack [7]. MFCs spontaneously generate
electrical current by pairing a bio-anode, on which exoelectrogenic microbes oxidize soluble
wastewater organics and release electrons [8], and cathodes with oxygen reduction [9]. The
entropic energy released by mixing solutions of different ionic strength is converted in a RED
stack into electrical energy by separating chambers of high concentration (HC) and low
concentration (LC) saline solutions with alternating anion (AEM) and cation (CEM) exchange
membranes [10-12]. The additive electrochemical junction potential (typically 0.1 – 0.2 V per cell
pair) [13] of the electrolytic-pile provides additional driving force for current generation at the
124
MFC electrodes. Power generation in MRCs is synergistic because the RED stacks not only
directly contribute voltage to power production but also enhance MFC electrode performance [7].
When the electrodes were transferred from a single-chamber MFC to a MRC with an 11
membrane RED stack fed AmB solutions, power production increased by >300% using acetate
and by >700% with domestic wastewater [14].
The improved performance of the MRC relative to a MFC is due to the use of multiple pairs
of membranes in the RED stack, the use of AmB in the stack, and the improved electrode
potentials. Power production with an 11 membrane stack with acetate reached 5.6 W/m2 based on
projected cathode area (or equivalently cross sectional area between the electrodes), compared to
1.1 W/m2 using only the MFC. While it is typical to normalize power density based on projected
cathode area for MFCs, the total membrane area is typically reported in RED studies. When
evaluated on the basis of total membrane area, MRC power densities have ranged from 0.3 to 0.5
W/m2 [7, 14]. The use of these membranes in the MRC adds substantial capital costs for reactor
construction compared to MFCs. Thus, reducing the number of membranes pairs or further
increasing power production could make MRCs a more commercially viable biotechnology.
The effect of AmB on MRC performance relative to that of NaCl has not been well
examined. AmB solutions can be prepared to create larger salinity gradients than those possible
with river water and seawater. However, there are other important differences that have not been
previously addressed. A HC solution with NaCl in the catholyte improves conductivity, but it
does not buffer pH changes. As a result, the cathode pH increases due to the consumption of
protons by the cathode and accumulation of OH–, leading to potential losses of >0.3 V [15-17].
Phosphate or carbonate buffers can be used to mitigate pH changes, but comparisons to high
conductivity NaCl solutions have shown that these negatively-charged buffer species do not
improve performance [17]. The use of AmB as the catholyte may have advantages compared to
these other chemical species because positively charged ammonium ions would be attracted to the
125
cathode, which could greatly affect oxygen reduction kinetics as well as hydroxide ion gradients
near the electrode. Since AmB has only been used as a catholyte in the presence of an 11
membrane RED stack [14], the contribution of improved cathode kinetics to MRC electrode
performance has not been separately examined from that of the stack performance.
The objectives of this study were to investigate potential catalytic benefits AmB on oxygen
reduction, and determine to what extent MRC performance could be improved using only a
minimal number of RED cell pairs. To study the effects of the AmB and cell pairs on MRC
performance, power production and internal resistance were examined using three different
reactor configurations: (1) a single chamber MFC with a bicarbonate buffer used as the single
electrolyte in contact with both electrodes; (2) a one cell pair (1-CP) MRC where the anolyte
formed an equivalent LC chamber with the bicarbonate buffer, and the catholyte was the HC
solution (1 M AmB); and (3) a two cell pair (2-CP) MRC containing an additional membrane pair
containing AmB LC and HC solutions (Figure 6.1). To investigate the effect of anolyte
composition on electrode and RED stack performance, sodium acetate (in 50 mM bicarbonate
buffer) and domestic wastewater were used as fuels. The effect on anolyte conductivity was
further examined in the 1-CP MRC configuration by amending the domestic wastewater with
bicarbonate buffer (50 mM). To quantify the specific effects of AmB catholyte and RED stack
potential on electrode activity, internal resistance was measured using linear polarization and
galvanostatic electrochemical impedance spectroscopy.
6.3 Materials and Methods
6.3.1 Reactor Construction
MFCs and MRCs were constructed as previously described [7] with minor
modifications. MFCs had a single 4-cm diameter chamber (Lexan, 28 mL empty bed
126
volume) (Figure 6.1). To maintain the same electrode spacing in both MFC and MRC
tests, the anode and cathode chambers of 1- and 2-CP MRCs were constructed using two,
2-cm long cubic chambers (14 mL empty bed volume). The anodes were graphite fiber
brushes (2.7 cm in diameter, 2.3 cm in length; Mill-Rose Labs Inc., OH) that were heat
treated [18] prior to being inoculated with effluent from an existing MFC, and enriched in
single chamber MFCs. The same anodes were used in all three reactor configurations.
The air-cathodes contained a Pt catalyst (0.5 mg Pt/cm2) on carbon-cloth, and were
constructed as previously described [19] with a Nafion catalyst binder on the water side,
and four layers of polytetrafluoroethylene (PTFE) diffusion layers on the air side.
The 1-CP MRC contained a single anion exchange membrane (1-CP; Selemion
AMV, Asahi Glass, Japan, 7 cm2 active area) separating the anode and cathode chambers
(Figure 6.1). The 2-CP MRC reactor contained an additional HC/LC cell pair between the
anode and cathode chambers, requiring three membranes (1 cation- and 2 anion-
exchange membranes, Selemion CMV and AMV, Asahi glass, Japan). Inter-membrane
chambers were sealed and separated by silicon gaskets (0.5 mm), each with an 8 cm2 (2
cm × 4 cm) rectangular cross section cut out. The total ionic exchange membrane area in
the 2-CP MRC was 24 cm2. The HC solution entered the reactor at the cathode and exited
from the HC cell next to the anode chamber. The LC AmB stream entered the RED stack
near the anode and flowed upward through the LC cell in the stack. Both the HC and LC
AmB solutions were continuously fed using a peristaltic pump (Cole Parmer, IL) at a
flow rate of 1.0 mL/min. During polarization tests fresh saline solutions were pumped
through the RED stack with the effluent collected in separate reservoirs. In batch recycle
experiments 0.1 L of each solution was recycled through the stack in airtight flow paths.
127
Before each fed-batch experiment the stack and tubing were flushed with matching
solutions.
6.3.2 Solutions
Anodes were fed either a buffered sodium acetate solution, or raw domestic
wastewater except as noted. The bicarbonate buffered acetate solution was designed to
reduce solution and membrane resistances and provide a high concentration of electron
donor for the anodic biofilm. The acetate solution contained 2.0 g/L of sodium acetate in
the 50 mM bicarbonate buffer (4.2 g/L NaHCO3) as well as 0.23 g/L NH4H2PO4, 0.16
g/L KCl, and trace vitamins and minerals [20]. The volume of acetate was 30 mL for
each reactor configuration. In the 1-CP and 2-CP configurations, the reduced anode
volume required recirculation of 16 mL of sodium acetate solution. Domestic wastewater
(WW) samples were collected from the Penn State University wastewater treatment plant
primary clarifier effluent. High conductivity wastewater (WW-HC) was prepared by
dissolving 4.2 g/L of NaHCO3 in the wastewater. To avoid electron donor limitations that
could occur due to low COD, experiments using wastewater were conducted by recycling
250 mL of substrate through the anode chamber of each reactor. The anolyte feed
reservoirs were airtight glass vessels submerged in ice water to minimize external organic
degradation.
The HC ammonium bicarbonate solution used in the 1- and 2-CP MRC reactors was
prepared by dissolving chemical grade ammonium bicarbonate salt (Alfa Aesar, MA) into
deionized water within an airtight vessel. The initial HC (1.0 M) and LC (0.01 M)
concentrations tested were selected based on previous optimization experiments [14].
128
6.3.3 Electrochemical Measurements and Analysis
Power densities and bulk internal resistances for each reactor configuration were
determined using chronopotentiometry. Whole cell current intervals (20 minutes) were
set with a galvanostat (Biologic VMP-3, USA). Prior to galvanostatic operation, reactors
were held at open circuit for 30 minutes.
Power density was calculated by multiplying the electrical current (i) and total cell
voltage. MFC electrode power contribution (PMFC) was determined by measuring the
anode potential (Van) and cathode potential (Vcat) against Ag/AgCl (+0.210 V vs SHE)
reference electrodes (BASi, IN), and calculated as PMFC = i(Vcat – Van).[14] Power
contributed by the RED stack was calculated as PRED = iVRED based on the stack voltage
(VRED), which was measured using two reference electrodes located on either side of the
stack.
MRC internal resistance (RMRC) is the sum of potential dissipative forces at the anode
(Ran), cathode (Rcat), and within the RED stack (Rstk), or RMRC = Ran + RRED + Rcat. Internal
resistance can be further separated into solution, membrane, kinetic, diffusion and
capacitive resistances. The solution and membrane resistance (Rsol+m) is dependent on
electrode spacing, electrolyte and membrane conductivity, as well as diffusion area.
Kinetic charge transfer occurs during current generation at the anode (Ran,ct) and current
consumption at the cathode (Rcat,ct). Diffusion resistance can arise due to transport of
substrate to the anode (Ran,diff) as well as protons and oxygen to the cathode (Rcat,diff).
Diffusion boundary layer resistance (RRED,dbl), also known as concentration polarization,
is caused by concentration and depletion of ions at the interface of ion-exchange
129
membranes. The fixed charges of IEMs also create ionic double layer capacitance
(RRED,dl) that impede ion diffusion within the RED stack [21-23].
Bulk internal resistances of the whole cell, anode, cathode and RED membrane
interfaces were determined from the slopes of linear polarization curves constructed for
each reactor component [22, 24]. Galvanostatic EIS (GEIS) was used to characterize the
effect of RED stacks on specific resistances. GEIS was conducted at a set current of 1.5
mA and amplitude of 0.21 mA. GEIS spectra were simultaneously recorded for each
reactor component by using three galvanostat channels operated in unison (one master
channel that applied AC current and two reference channels that collected EIS spectra).
Whole cell impedance was collected from the master channel terminals by setting the
cathode as the working electrode and the anode as both the counter and reference
electrode. The two reference channels were used to obtain anode, RED, and cathode
impedance spectra. Anode and cathode impedance was measured against an Ag/AgCl
reference electrode. RED stack impedance was obtained with reference electrodes
positioned at either side of the membrane stack. All EIS experiments were conducted
within one hour of chronopotentiometry measurements to ensure similar working
conditions. EIS data were analyzed using EC-Lab V10.32 to determine the specific
resistances of each reactor component. Equivalent circuit model (ECM) software was
used to determine specific resistances from the Nyquist plots of each reactor component
(Figure A3.5).
During batch-fed experiments, the voltage drop across an external resistor was
recorded every 10 minutes by a digital multi-meter (Keithley Instruments, OH). To
maximize energy production, the fixed external resistance was set to match the internal
130
resistance of the cell at each condition as determined from polarization data. All batch
recycle experiments were conducted at 30 °C in a constant temperature room.
Elelctrical energy recovery, in batch experiments with recycled anolyte, was
calculated as the energy produced (Wh) from the substrate based on chemical oxygen
demand removed (g-COD). Power production (W) was calculated for each measured
voltage (V2/Rext), multiplied by the sample interval (h) and summed over the entire fed-
batch cycle. Volumetric COD removal rates were normalized by the duration of the fed-
batch cycle and the anode (an) volume (g-COD/Lan-d) to allow direct comparison to
commercialized technologies. COD removal was determined as previously described
[25].
6.4 Results and Discussion
6.4.1 MRC Power Production
The use of the concentrated AmB solution as the catholyte in the 1-CP MRC (two
chamber) configuration resulted in a >300% increase in power (3.1 ± 0.1 W/m2)
compared to that obtained the single-chamber MFC containing only bicarbonate buffer
(0.9 ± 0.05 W/m2-cat;). The 1-CP MRC also produced higher power with domestic
wastewater (1.8 ± 0.03 W/m2) than the MFC (0.3 ± 0.02 W/m2-cat). Previously, power
densities produced using two-chambered MFCs with concentrated NaCl or a phosphate
buffer as catholytes, have always been lower those obtained with single-chamber MFCs
[15, 17]. The use of concentrated AmB catholyte in the two-chamber configuration (1-
CP) reduced several limitations compared to two-chamber MFCs for bio-electricity
production, that include: transport of HCO3– into the anode chamber instead of Cl–
131
buffered anode pH and prevented acidification; the presence of NH4+ near the cathode
mitigated accumulation of OH-; a favorable concentration gradient that created a positive
membrane potential and minimized power loss due to membrane resistance [17].
The inclusion of a second RED cell pair between the electrodes (2-CP) further
increased the peak power density to 4.1 ± 0.1 W/m2-cat (1.3 ± 0.1 W/m2-mem, Figure
6.2a). Power density from the 2-CP reactor fed wastewater was only slightly higher (1.9
± 0.05 W/m2-cat, 0.6 ± 0.02 W/m2-mem, Figure 6.2b) than the 1-CP configuration. When
fed the dilute and complex organic matter in wastewater, bio-anodes did not sustain
negative potentials at current densities beyond 6.5 A/m2-cat (Figure 6.2d). At these
higher current densities, rapid anode polarization was observed. The low conductivity of
wastewater also significantly reduced the effectiveness of the RED stack.
6.4.2 Electrode and RED Stack Potential Losses in MRCs
At peak power production in both 1- and 2-CP MRCs, the RED stack potential was
nearly 0 mV or negative. Under these conditions, the major reason for the power increase
relative to MFC conditions was therefore a reduction in electrode potential loss at higher
current densities. Under direct current conditions (DC), electrode potential loss (mV/mA)
can be quantified as polarization resistance (Ω). During operation with the high
concentrate AmB electrolye, cathode potential loss with increased current was uniformly
reduced (relative to the MFC) in both the 1- and 2-CP reactor configurations (Figure
6.2c,d). The cathode linear polarization resistance in the single-chamber MFC was 55 Ω
(or 55 mV/mA) with buffered acetate, and 220 Ω with domestic wastewater. When
operated in HC AmB, the cathode polarization resistance dropped dramatically to 12 - 15
132
Ω in both the 1-CP and 2-CP MRC reactors, and was independent of anolyte
composition.
Direct current anode potential loss was influenced by the use of AmB catholyte as
well as the magnitude of RED stack potential. In the single chamber MFCs, anode
potential loss was highly dependent on the type of anolyte, with nearly a 90 Ω (or
mV/mA) difference between bicarbonate buffered acetate (33 Ω) and domestic
wastewater (120 Ω). In the 1-CP MRC, anode potential loss decreased significantly for
both substrates (Ac: 21 Ω, WW: 28 Ω). In the 2-CP MRC configuration, additional RED
stack voltage minimized anode DC resistance to the point at which potential loss was not
appreciably affected by substrate composition (10 Ω in WW versus 8 Ω in buffered
acetate, Figure S1). The similarity in 2-CP MRC anode potential loss implies that the
magnitude of RED potential may influence bio-anode activity.
Electrode potentials, and the rate of potential loss, in the 1-CP and 2-CP MRCs were
nearly identical with both organic fuel solutions, but power production from acetate
remained significantly higher than from wastewater. This disparity can be linked to the
effect that substrate composition had on RED stack resistance. With buffered acetate in
the anode chamber, RED resistance was 17 Ω in the 1-CP reactor, and it increased to 36
Ω with the inclusion of an additional cell pair (Figure 4b). When domestic wastewater
filled the anode chamber, the RED stack resistance significantly increased in both the 1-
CP (38 Ω) and 2-CP (97 Ω) MRCs. In order to increase power production from low
conductivity wastewater, transport resistance at the terminal AEM/anolyte interface must
be reduced. Minimizing anode chamber thickness would reduce solution resistance and
133
increasing solution flow rate at the membrane surface could reduce boundary layer
resistance [23, 26].
6.4.3 Anolyte Composition and Membrane Transport Resistance
The two anode substrates evaluated in this study had different characteristics in terms
of electrode donor concentration and composition, buffer capacity, and specific
conductance. In order to understand how buffer concentration influenced the specific
resistances limiting MRC power production, experiments using GEIS were conducted on
the individual reactor components (anode, RED interface, and cathode) with the same set
current (1.5 mA) and amplitude (0.21 mA), using three different anode substrates:
buffered acetate; raw domestic wastewater (WW); and wastewater amended with 4.2 g/L
NaHCO3 (WW-HC) that matched the conductivity of the buffered acetate.
The differences between membrane interface impedance values in the 1-CP MRC
were primarily due to anolyte conductivity. Membrane resistance determined by
alternating current (AC) impedance spectroscopy showed ohmic resistance (from anode
and cathode solutions as well as the AEM, Rsol+m) and two overlapping semicircles
representing double layer capacitance (Rdl) and diffusion boundary layer (Rdbl) resistance
(Figure A3.3c). All three types of membrane impedance were highest when the MRC was
fed raw domestic wastewater (Rsol+m: 8 ± 0.1 Ω, Rdl: 15 ± 1.3 Ω, Rdbl: 17 ± 1.1 Ω, Figure
6.3b). Raising the conductivity of the wastewater decreased both double layer and
diffusion boundary layer resistance (Rsol+m: 6.0 ± 0.1 Ω, Rdl: 6.0 ± 0.4 Ω, Rdbl: 12 ± 0.2 Ω,
Figure 6.3b). By raising anolyte conductivity and reducing membrane transport
resistance, power density of the 1-CP MRC fed high conductivity wastewater (2.1 ± 0.05
134
W/m2, Figure S2a) exceeded that of the 2-CP MRC with raw domestic wastewater.
However, raising wastewater conductivity did not reduce membrane impedance to the
levels of buffered acetate (Rsol/m: 5.0 ± 0.3 Ω, Rdl: 4.0 ± 0.1 Ω, Rdbl: 7.0 ± 0.4 Ω). This
implies that some other distinguishing characteristic of the wastewater, such as presence
of particulate organic matter or other ions, may have adversely affect membrane
capacitance and boundary layer resistance. The summation of individual membrane
resistances determined by GEIS closely corresponded with linear polarization resistance
determined under direct current (DC) conditions (differences ranged from 0.5 − 2.8 Ω).
Electrode resistance values determined by GEIS were nearly identical for each
substrate condition. Alternating current (AC) cathode impedance included 3.7 ± 0.4 Ω of
solution resistance and 12 ± 0.5 Ω of charge transfer resistance for all three conditions
(Figure 3b, S3c). The total cathode impedance values, based on the sum of individual AC
resistances, were all quite similar (within 2 Ω) to cathode resistance values determined by
the linear polarization slope method (DC resistance). Anode impedance spectra for all
three conditions were comprised of a single charge transfer semicircle (Figure S3a).
Anode charge transfer resistance values ranged from 13 − 15 Ω (Figure 3b, S3a) and
were lower than DC resistances for all three substrates (Figure 3a,b). The discrepancy
between DC (linear polarization) and AC (GEIS) anode resistance (WW: 15 Ω, WW-HC:
13 Ω, Ac: 7 Ω) decreased with the magnitude of RED stack resistance (WW: 40 Ω, WW-
HC: 24 Ω, Ac: 16 Ω) implying that 1-CP anode potential, under direct current conditions,
may have been adversely affected by the negative potential of the membrane interface.
135
6.4.4 Effect of RED Stack Potential on Bio-Anode Charge Transfer Resistance
Anode resistances, determined from linear polarization experiments (DC resistance),
sharply decreased as internal resistance decreased (1-CP) and RED stack potential
increased (2-CP). GEIS was used to track anode biofilm catalytic activity, represented by
charge transfer resistance, in the single chamber MFC, 1-CP and 2-CP MRCs. In each
configuration, the anode was fed buffered acetate and GEIS was conducted as the same
set current (1.5 mA) and amplitude (0.21 mA).
Nyquist plots of anode impedance in the single chamber MFC showed ohmic solution
resistance (Rsol: 12 ± 0.1 Ω) between anode and reference as well as charge transfer and
diffusion resistance (Rct: 16 ± 0.01 Ω, Rdiff: 11 ± 0.7 Ω, Figure 6.4b, S4a). Anode
impedance in the 1-CP configuration was represented by a single charge transfer
semicircle that was similar, based on radius (Rct: 14 ± 0.2 Ω, Figure 6.4b, S4a), to the
anode kinetic resistance in the single chamber MFC. This similarity in charge transfer
implies that eliminating solution and diffusion resistance in the 1-CP reactor led to the
observed reduction in anode potential loss. In the 2-CP reactor, anode charge transfer
resistance decreased to 4.2 ± 0.8 Ω. At the GEIS set current, the 1-CP reactor RED
potential was 0.005 mV and 2-CP RED potential was 150 mV, implying that positive
RED potential not only adds to MRC power, but also enhances the catalytic activity of
anodic biofilms.
6.4.5 Wastewater Treatment Rates and Energy Recovery
To investigate how increased MRC power would translate to COD removal and
energy recovery, additional tests were conducted with buffered acetate or domestic
136
wastewater recycled through the anode chamber to maintain a similar amount of total
COD in the samples. The rate of COD removal with the 1-CP system with acetate was
1.5 ± 0.3 g-COD/Lan-d, nearly twice that of the MFC (Figure 6.6a). In the 2-CP MRC,
COD removal further increased to 1.9 ± 0.1 g-COD/Lan-d COD. Overall COD removal
(77 – 91%) was very high in both the 1-CP and 2-CP MRCs fed acetate. Energy recovery
increased from 0.4 ± 0.01 Wh/g-COD for the MFC, to 0.7 ± 0.01 Wh/g-COD for the 1-
CP and 0.9 ± 0.1 for the 2-CP MRCs (Figure 6.6b).
Volumetric COD removal rates and energy recoveries were also greatly increased
with domestic wastewater. The COD removal rate with the 1-CP MRC (1.2 ± 0.1 g-
COD/Lrxtr-d) was 20 times larger than that of the single chamber MFC (0.05 ± 0.007 g-
COD/Lrxtr-d, Figure 6.6a). Using the 2-CP MRC, the COD removal rate of 2.3 ± 0.2 g-
COD/Lrxtr-d was nearly twice that of the 1-CP, and it was actually larger than the acetate
removal rate in this system. This COD removal rate for wastewater is within the range
reported for activated sludge systems [27]. The overall COD removal with the MRCs (1-
CP, 40 ± 6%, 2-CP, 49 ± 1%) was lower than that obtained with the single chamber MFC
(68 ± 2%). Energy recovery with wastewater followed the same trend as acetate, but
values were lower due COD removal from processes other than oxidation by anode
bacteria (Figure 6.6b). Coulombic efficiencies, which measure the percentage of COD
removed by current generation, were significantly lower with wastewater (35 – 50%)
than with acetate (72 – 81%).
137
6.5 Outlook
By using AmB as catholyte and minimizing the number of RED cell pairs, it was
shown that electrode kinetics could be dramatically enhanced with only a single
membrane. Although adding additional RED cell pairs could further increase power
densities, calculated on the basis of projected cathode area, power does not proportionally
increase on the basis of total membrane area. The inclusion of additional membranes
lowered the membrane area normalized power density from 3.1 ± 0.1 W/m2 in the 1-CP
configuration to 1.4 ± 0.1 W/m2. These values are significantly greater than membrane
area normalized power densities of 5-CP MRCs (0.3 — 0.5 W/m2) and quite comparable
to those reported for optimized abiotic RED system.[7, 14, 26, 28]
In order to increase the MRC performance on the basis of total membrane area, the
resistance per RED cell pair must be decreased. The electrode chamber and RED flow
channel thickness used in this study promoted high solution and transport boundary layer
resistance, leading to 17 Ω per CP with acetate and 48 Ω per CP with wastewater.
Previous studies have shown that optimization of RED flow channel thickness and
hydrodynamics can reduce RED stack resistance to less than 1 Ω per CP.[26, 29] If MRC
stack resistance can attain similar optimization, 2-CP power densities would approach 10
W/m2-cathode (3 W/m2-membrane, Figure 6.8a, b). With stack optimization, 2-CP MRCs
could exceed the maximum membrane normalized power densities reported for abiotic
RED systems (2.2 W/m2) [26] showing that increasing MFC electrode power with
minimal RED stacks makes efficient use of membrane area and could enable high power
densities even using domestic wastewater.
138
6.6 Tables
Table 6.1: Anode substrate characteristics
Anolyte tCOD (g/L) sCOD (g/L) κ (mS/cm) pH
Ac 1.8 ± 0.05 1.8 ± 0.05 5.5 ± 0.1 8.1 ± 0.1
WW 0.38 ± 0.04 0.20 ± 0.02 1.5 ± 0.3 7.6 ± 0.2
WW-HC 0.38 ± 0.04 0.20 ± 0.02 5.1 ± 0.1 8.1 ± 0.1
139
6.7 Figures
Figure 6.1: Schematics of tested reactor configurations.
Single Chamber MFC
Ref.
2-CP MRC(Additional HC/LC)
AEM
CEMRef. Ref.
1-CP MRC(HC Cathode)
Ref. Ref.
AEM
-0.2
-0.1
0.0
0.1
0.2
0.3
0 2 4 6 8 10 12
RED
Sta
ck P
oten
tial (
V)
Current Density (A/m2)
2-CP Ac
1-CP Ac
-0.2
-0.1
0.0
0.1
0.2
0.3
0 1 2 3 4 5 6 7
RE
D S
tack
Pot
entia
l (V
)
Current Density (A/m2 )
1-CP WW
2-CP WW
Figure 6.2: (a,b) Power densities normalized to cathode area, (c,d) anode and cathode potentials (vs. SHE), and (e,f) reverse electrodialysis (RED) stack potentials of MFCs, 1-CP and 2-CP MRCs fed (filled markers) bicarbonate buffered acetate and (open markers) domestic wastewater.
-0.4
-0.2
0.0
0.2
0.4
0.6
0 2 4 6 8 10 12
Ele
ctro
de P
oten
tial (
V)
Current Density (A/m2)
2-CP Ac C 2-CP Ac A1-CP Ac C 1-CP Ac AMFC Ac C MFC Ac A
-0.4
-0.2
0.0
0.2
0.4
0.6
0 1 2 3 4 5 6 7
Ele
ctro
de P
oten
tial (
V)
Current Density (A/m2)
2-CP WW C 2-CP WW A1-CP WW C 1-CP WW AMFC WW C MFC WW A
c)
d)
e)
f)
0.0
1.0
2.0
3.0
4.0
5.0
0 2 4 6 8 10 12
Pow
er D
ensi
ty (W
/m2 -
cat)
Current Density (A/m2)
MFC Ac1-CP Ac2-CP Ac
0.0
0.5
1.0
1.5
2.0
2.5
0 1 2 3 4 5 6 7
Pow
er D
ensi
ty (W
/m2 -
cat)
Current Density (A/m2)
MFC WW1-CP WW2-CP WW
b)
a)
140
0
20
40
60
80
100
1-CP WW 1-CP HKWW
1-CP Ac
DC
Inte
rnal
Res
ista
nce
(Ω) Cathode
REDAnode
0
20
40
60
80
100
1-CP WW 1-CP HKWW
1-CP Ac
AC
Inte
rnal
Res
ista
nce
(Ω)
Cat_Rct Cat_RsolRED_Rdbl RED_RdlRED_Rs/m An_Rct
Figure 6.3: Internal resistance of 1-CP reactors components fed raw domestic wastewater (WW), buffered wastewater (WW-HC), and buffered acetate (Ac), determined from (a) direct current (DC) polarization and (b) alternating current galvanostatic electrochemical impedance spectroscopy.
a)
0
20
40
60
80
100
MFC Ac 1-CP Ac 2-CP Ac
AC
Inte
rnal
Res
ista
nce
(Ω) Cat_Rct Cat_Rsol
RED_Rdbl RED_RdlRED_Rs/m An_RsolAn_Rdiff An_Rct
0
20
40
60
80
100
MFC Ac 1-CP Ac 2-CP Ac
DC
Inte
rnal
Res
ista
nce
(Ω) Cathode
REDAnode
b)
b) a)
Figure 6.4: Internal resistance of MFC, 1-CP, and 2-CP MRC reactors components fed buffered acetate (Ac) determined from (a) direct current (DC) polarization and (b) alternating current (AC) galvanostatic electrochemical impedance spectroscopy.
141
Figure 6.5: a) Volumetric COD removal rates and (b) energy recovery from wastewater and acetate in batch recycle experiments.
b)
b)
a)
142
0
1
2
3
4
5
6
0 25 50 75 100(W
/m2 -
mem
bran
e)
0
2
4
6
8
10
12
0 25 50 75 100
Pow
er D
ensi
ty (W
/m2 -
cat)
Reduction in RED Resistance (%)
2-CP Ac2-CP WW1-CP Ac1-CP WW
Figure 6.6: Estimation of 1-CP and 2-CP MRC power density with buffered acetate (Ac) and wastewater (WW) normalized to (a) total membrane area and (b) cathode area versus reduction in RED stack resistance.
b)
143
6.8 Literature Cited
1. WIN, Clean & Safe Water for the 21st Century: A Renewed National Commitment to
Water and Wastewater Infrastructure, 2000.
2. IPCC, Climate Change 2007: Synthesis Report. , in Contribution of Working Groups I, II
and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate
Change2007, IPCC: Geneva.
3. McCarty, P.L., J. Bae, and J. Kim, Domestic Wastewater Treatment as a Net Energy
Producer–Can This be Achieved? Environmental Science & Technology, 2011. 45(17):
p. 7100-7106.
4. McGinnis, R.L., J.R. McCutcheon, and M. Elimelech, A novel ammonia-carbon dioxide
osmotic heat engine for power generation. Journal of Membrane Science, 2007. 305: p.
13-19.
5. EIA, U.S., Annual Energy Review 2010, DOE, Editor 2010.
6. Kamat, P.V., Meeting the clean energy demand: Nanostructure architectures for solar
energy conversion. The Journal of Physical Chemistry C, 2007. 111(7): p. 2834-2860.
7. Kim, Y. and B.E. Logan, Microbial Reverse Electrodialysis Cells for Synergistically
Enhanced Power Production. Environmental Science & Technology, 2011. 45(13): p.
5834-5839.
8. Logan, B.E., Exoelectrogenic bacteria that power microbial fuel cells. Nature Reviews
Microbiology, 2009. 7(5): p. 375-381.
9. Liu, H., R. Ramnarayanan, and B.E. Logan, Production of electricity during wastewater
treatment using a single chamber microbial fuel cell. Environmental Science &
Technology, 2004. 38(7): p. 2281-2285.
10. Wick, G.L., Power from salinity gradients. Energy, 1978. 3(1): p. 95-100.
144
11. Post, J.W., et al., Salinity-gradient power: Evaluation of pressure-retarded osmosis and
reverse electrodialysis. Journal of Membrane Science, 2007. 288(1-2): p. 218-230.
12. Ramon, G.Z., B.J. Feinberg, and E.M.V. Hoek, Membrane-based production of salinity-
gradient power. Energy & Environmental Science, 2011.
13. Weinstein, J.N. and F.B. Leitz, Electric power from differences in salinity: the dialytic
battery. Science, 1976. 191(4227): p. 557.
14. Cusick, R.D., Y. Kim, and B.E. Logan, Energy Capture from Thermolytic Solutions in
Microbial Reverse-Electrodialysis Cells. Science, 2012. 335(6075): p. 1474-1477.
15. Kim, J.R., et al., Power generation using different cation, anion and ultrafiltration
membranes in microbial fuel cells. Environmental Science & Technology, 2007. 41(3): p.
1004-1009.
16. Popat, S.C., et al., Importance of OH− Transport from Cathodes in Microbial Fuel Cells.
ChemSusChem, 2012. 5(6): p. 1071-1079.
17. Ahn, Y. and B.E. Logan, Saline catholytes as alternatives to phosphate buffers in
microbial fuel cells. Bioresource Technology, 2013. 132(0): p. 436-439.
18. Wang, X., et al., Use of carbon mesh anodes and the effect of different pretreatment
methods on power production in microbial fuel cells. Environmental Science &
Technology, 2009. 43(17): p. 6870-6874.
19. Cheng, S., H. Liu, and B.E. Logan, Increased performance of single-chamber microbial
fuel cells using an improved cathode structure. Electrochemistry Communications, 2006.
8(3): p. 489-494.
20. Ambler, J.R. and B.E. Logan, Evaluation of stainless steel cathodes and a bicarbonate
buffer for hydrogen production in microbial electrolysis cells using a new method for
measuring gas production. International Journal of Hydrogen Energy, 2011. 36(1): p.
160-166.
145
21. Zhang, F., et al., Mesh optimization for microbial fuel cell cathodes constructed around
stainless steel mesh current collectors. Journal of Power Sources, 2011. 196(3): p. 1097-
1102.
22. Hutchinson, A.J., J.C. Tokash, and B.E. Logan, Analysis of carbon fiber brush loading in
anodes on startup and performance of microbial fuel cells. Journal of Power Sources,
2011. 196(22): p. 9213-9219.
23. Długołęcki, P., et al., On the resistances of membrane, diffusion boundary layer and
double layer in ion exchange membrane transport. Journal of Membrane Science, 2010.
349(1): p. 369-379.
24. Logan, B.E., Microbial fuel cells2008, Hoboken, NJ: John Wiley & Sons, Inc. 300.
25. Liu, H., S. Cheng, and B.E. Logan, Power generation in fed-batch microbial fuel cells as
a function of ionic strength, temperature, and reactor configuration. Environmental
Science & Technology, 2005. 39(14): p. 5488-5493.
26. Vermaas, D.A., M. Saakes, and K. Nijmeijer, Doubled power density from salinity
gradients at reduced intermembrane distance. Environmental Science & Technology,
2011. 45(16): p. 7089-7095.
27. Rozendal, R.A., et al., Towards practical implementation of bioelectrochemical
wastewater treatment. Trends in Biotechnology, 2008. 26(8): p. 450-459.
28. Vermaas, D.A., M. Saakes, and K. Nijmeijer, Power generation using profiled
membranes in reverse electrodialysis. Journal of Membrane Science, 2011. 385–386(0):
p. 234-242.
29. Długołȩcki, P., et al., Practical Potential of Reverse Electrodialysis As Process for
Sustainable Energy Generation. Environmental Science & Technology, 2009. 43(17): p.
6888-6894.
146
Appendix A
Energy Efficient Phosphate Recovery as Struvite Within a Single Chamber Microbial Electrolysis Cell
Figure A.1: Concurrent LSV (filled) and pH (open) measurements at the surface of SSM cathode in 3mM PBS and 75 mM HCO3
- electrolyte.
-30
-25
-20
-15
-10
-5
0
6.5
7.0
7.5
8.0
8.5
9.0
-0.1 -0.4 -0.7 -1.0
I (m
A)pH
E (V) vs. SHE
pH, 3mM PBS
pH, 75 mM HCO3
I (mA), 3 mM PBS
I (mA), 75 mM HCO3
147
Figure A.2: Average batch time and crystal accumulation of MESC reactors with both SSF and SSM cathodes at applied voltages of 0.75, 0.90 and 1.05 V.
10
20
30
40
0.0
2.0
4.0
6.0
8.0
10.0
0.7 0.8 0.9 1.0 1.1
Bat
ch D
urat
ion
(hr)
Cry
stal
l Gro
wth
(mg/
batc
h)
Applied Voltage (V)
SSMSSFBatch
2
A.3: Anode (an) and cathode (cat) potentials measured in fed-batch operation at a) 0.75 b) 0.90 and c) 1.05 V.
-1.2
-1.0
-0.8
-0.6
-0.4
-0.2
0.00.7 0.8 0.9 1.0 1.1
Ele
ctro
de P
ot.
(V) v
s. S
HE
Applied Voltage (V)
SSM-C-cat
SSM-S-cat
SSF-C-cat
SSF-S-cat
SSM-C-an
SSM-S-an
SSF-C-an
SSF-S-an
148
Figure A.4: EIS determination of internal resistance for SSM and SSF in MEC (C) and MESC (S) reactors at cathode potentials observed in fed-batch operation at a) 0.75 b) 0.90 and c) 1.05 V.
a)
b)
c)
Appendix B
Energy Production from Thermolytic Solutions in Microbial Reverse Electrodialysis Cells
1
Figure B.1: Relationship between ammonium bicarbonate concentration and solution. Figure
y = -16.505x2 + 91.209xR² = 1
0
20
40
60
80
100
120
140
0.0 0.5 1.0 1.5 2.0
Con
duct
ivity
(mS
/cm
)
Concentration (M)
150
Figure B.2: Relationship between total ammonium bicarbonate salt concentration and species concentration at pH =7 and T = 25 °C as estimated with OLI stream analyzer software.
0.0
0.5
1.0
1.5
2.0
0.0 0.5 1.0 1.5 2.0
Spe
cies
Con
cent
ratio
n (M
)
Total Concentration (M)
NH4+HCO3-NH4CO3-CO32-NH3
Figure B.3: Relationship between species concentration and species activity at pH =7 and T = 25 °C as estimated with OLI stream analyzer software.
0.0
0.3
0.5
0.8
1.0
0.0 0.5 1.0 1.5 2.0
Spe
cies
act
vity
Species Concentration (M)
NH4+HCO3-NH4CO3-CO32-
151
0.0
1.0
2.0
3.0
4.0
5.0
6.0
0.0 0.2 0.4 0.6 0.8 1.0P
ower
Den
sity
(W/m
2 )
Current Density (mA/cm2)
1.6 mL/min0.85 mL/min
Figure B.4: Power density curves of the MRC (HC = 0.95 M, SR = 100) at different salt solution flow rates.
1
Figure B.5: Batch recycle component (MFC, RED and total MRC) power profile of MRC fed sodium acetate and operating with an external resistance of 300 Ω.
0.0
1.0
2.0
3.0
4.0
0.0 0.2 0.4 0.6 0.8
Pow
er (W
/m2 )
Time (days)
MRCMFCRED
152
Appendix C
Minimal RED Cell Pair Microbial Reverse Electrodialysis Cells
Figure C.1: Internal resistance determined from the slope of linear polarization curves for the MFC, 1-CP and 2-CP MRC fed acetate (Ac) and domestic wastewater (WW).
050
100150200250300350
MFCWW
MFCAc
1-CPWW
1-CPAc
2-CPWW
2-CPAc
Inte
rnal
Res
ista
nce
(Ω)
CathodeREDAnode
153
a) b)
Figure C.2: a) Power density, (b) electrode potentials, and (c) stack potentials of 1-CP reactor fed acetate (Ac), high conductivity wastewater (WW-HC) and raw wastewater (WW).
) ) )
Figure C.3: Nyquist plots of GEIS data collected from 1-CP MRCs fed acetate, wastewater, and high conductivity wastewater. Impedance spectra were simultaneously collected for (a) whole cell and (b) anode (c) RED stack and (d) cathode.
b) d)
a) c)
b) d)
154
0
10
20
30
40
0 10 20 30 40
-Img
(Z)/
Ohm
Real(Z)/Ohm
Cathode 2-CP Ac
Cathode 1-CP Ac
Cathode MFC
0
10
20
30
40
0 10 20 30 40
-Img
(Z)/
Ohm
Real(Z)/Ohm
RED 2-CP AcRED 1-CP Ac
05
10152025303540
0 10 20 30 40
-Img
(Z)/
Ohm
Real(Z)/Ohm
Anode MFC
Anode 1-CP Ac
Anode 2-CP Ac
Figure C.4: Nyquist plots of GEIS data collected from electrode and RED of MFC and MRCs fed acetate. Impedance spectra were simultaneously collected for (a) anode (c) RED stack and (d) cathode.
a) b) c))
0
5
10
15
20
0 5 10 15 20
-Img
(Z)/
Ohm
Real(Z)/Ohm
Anode 1-CP Ac
Anode model fit
0
5
10
15
20
0 5 10 15 20
-Img
(Z)/
Ohm
Real(Z)/Ohm
RED 1-CP Ac
Stack model fit
0
5
10
15
20
0 5 10 15 20
-Img
(Z)/
Ohm
Real(Z)/Ohm
1-CP Ac Cat
Cathode model fit
Figure C.5: Nyquist plots with equivalent circuit models and fits of GEIS data collected from the (a) anode (c) RED stack and (d) cathode of 1-CP MRC fed acetate.
a) b) c))
v
Rdl
>>Qa
Rsol+m v
Rdbl
>>Qa
v
Rct
>>Qa
Rsolv
Rct
>>Qa
Rsol
VITA
Roland D. Cusick
EDUCATION
Doctor of Philosophy in Environmental Engineering, Department of Civil and Environmental Engineering, The Pennsylvania State University, University Park, PA, 08/2010 – 12/2013 (expected). Advisor: Bruce E. Logan
Master of Science in Environmental Engineering, Department of Civil and Environmental Engineering, The Pennsylvania State University, University Park, PA, 08/2008 – 08/2010. Advisor: Bruce E. Logan
Bachelor of Science in Environmental Engineering, Department of Chemical and Environmental Engineering, The University of California, Riverside, CA, 09/2001–12/2005. AWARDS
• W. Wesley Eckenfelder Graduate Research Award, American Academy of Environmental Engineers and Scientists, 2013
• Alumni Association Dissertation Award, Penn State University, 2013
• DOW Sustainability Innovation Student Challenge Award, Grand Prize, Penn State University, 2012
• George W. Johnstone Graduate Fellowship, Department of Civil and Environmental Engineering, Penn State University, 2012
• Cecil Pepperman Memorial Fellowship, Department of Civil and Environmental Engineering, Penn State University, 2010
INVENTION DISCLOSURES AND PATENT APPLICATIONS
• M. Yates, Cusick, R. D., and B. E. Logan (2013). Formation of a Nanoporous Structure using an Active Exoelectrogenic Biofilm as a Template.
• Y. Kim, Cusick, R. D., and B. E. Logan (2012). Reverse Electrodialysis Supported Microbial Fuel Cells and Microbial Electrolysis Cells.
