1 nrnnrr11 Validation of

Solid Oxide Fuel Cell (SOFC) is a promising technology for producing electricity cleanly and efficiently. This type of fuel cell is a high temperature...

0 downloads 0 Views 5MB Size
Model Dev Samaria [

1nrnnrr11

· Validation of

0000086912

DC) Based

Solid Oxide Fuel Cell Operating with Practical Fuels

by

Mazni Ismail

A thesis presented to the University of Waterloo in fulfillment of the thesis requirement for the degree of Doctor of Philosophy m

Chemical Engineering

Waterloo, Ontario, Canada, 2013

© Mazni Ismail 2013

(1·11111111 .PERPUSTAKAAN UNIVERSiTI MALAYSIA PAHANG ~·~--------

No. Perolehan

No. Panggilan

086912

1~~\

Tarikh

' fi\"itl..

~ 0 I'S

0 1 JUL 2011;

'('

'\ °l\,U;)

Abstract Solid Oxide Fuel Cell (SOFC) is a promising technology for producing electricity cleanly and efficiently. This type of fuel cell is a high temperature fuel cell operating around 1000°C for state-of-the-art SOFC. An advantage of the high temperature is the possibility of combined heat and power generation which would even further increase the efficiency of this technology. However, due to high operating temperatures, there are problems associated with the development and commercialization of SOFC, such as requirement of high temperature gas seals, and relatively poor long-term stability. The current trend in SOFC development is therefore to reduce the operating temperature of the cell to the range 600-800°C. However, this requires developing new cell designs and materials since decreasing the operating temperature increases the ohmic overpotential due to higher ionic diffusion resistance in the electrolyte, thereby reducing electrochemical performance. For intermediate temperature SOFC, SDC is a promising electrolyte material to reduce the ohmic overpotential. The present research focused on developing a lD model of SDC based SOFC validated for a number of feed gas compositions, from humidified H2, mixture of CO and C02, to several syngas compositions (typical of diesel syngas, biomass syngas and pre-reformed natural gas). The model was developed for an anode supported cell. Few parameters were used as free fit parameters: essentially structural parameters, such as porosity and tortuosity, as well as kinetic parameters for H1 and CO electrochemical reactions. In most cases, the simulated results (polarization curve) fitted well the experimental data. It was seen that the performance of CO/C02 system is considerably lower than the H2/H20 system. The model results also allowed to access variables' profiles that would not be accessible experimentally, such species composition profile and local current density along the anode. In particular, it was observed that most the electrochemical reaction occurred within 10 µm away from the anode/electroIyte interface. In the literature, the water-gas shift (WGS) reaction is considered to occur only over Ni, but

the present work demonstrated that SDC is active toward the WGS reaction. Therefore, a kinetic study was carried out to determine a rate expression for the WGS reaction. This rate Ill

expression was then incorporated into the SOFC model. The results indicated that inclusion of the WGS reaction on SDC has minor or negligible effect in most situations, except in the case of CO mole fraction for the diesel syngas feed at higher cell voltage. The reason was that the composition of diesel syngas was such that there was a higher driving force for the WGS reaction to proceed in the reverse WGS direction. When the water content is high enough, as in the case of higher current densities, the form of the derived rate expression for the WGS on SDC makes the value of this rate very small. The rate expression was derived using relatively small amounts of water because of experimental limitation and therefore, the form of this rate needs to be revisited by considering higher amount of water.

IV

Table of Contents AUTHOR'S DECLARATION ............................................................................................................... ii Abstract ................................................................................................................................................. iii Acknowledgements ................................................................................................................................ v Dedication ............................................................................................................................................. vi Table of Contents ................................................................................................................................. vii List of Figures ....................................................................................................................................... ix List ofTables ........................................................................................................................................ xii Nomenclature ......................................................................................................................................x iii Chapter 1 Introduction ............................................................................................................................ 1 I. I Motivation of the Research .......................................................................................................... 3 1.2 Research Contributions ................................................................................................................ 4 I .3 Thesis Outline ............................................................................................................................... 4 Chapter 2 Literature Review .................................................................................................................. 6 2. I Introduction .................................................................................................................................. 6 2.2 Fuel Cells in Brief ........................................................................................................................ 6 2.3 Solid Oxide Fuel Cell ................................................................................................................. 10 2.4 Summary of SOFC Modelling Studies ....................................................................................... 23 Chapter 3 Experimental - Apparatus and Procedures ........................................................................... 35 3. I Material Preparation ................................................................................................................... 3 5 3 .1.1 Material Preparation for Methane Steam Reforming (MSR) Experimental Work .............. 35 3.1.2 Material Preparation for Reverse Water Gas Shift (RWGS) Reaction Experimental Work36 3 .2 Fixed Bed Reactor for Catalyst Performance Experiments ........................................................ 36 3.3 SOFC Button cell Electrochemical Performance Measurement ............................................... .42 Chapter 4 Kinetic Study of Reverse Water Gas Shift Reaction .......................................................... .49 4.1 Introduction ................................................................................................................................ 49 4.2 Methane Steam reforming on YSZ/Ni-YSZ and SDC/Ni-SDC ................................................. 49 4.3 Reverse Water Gas Shift Reaction on SDC and YSZ ................................................................ 55 4.4 Kinetics ofreverse Water Gas Shift Reaction on SDC .............................................................. 59 4.5 Markov Chain Monte Carlo Study ............................................................................................. 76 Chapter 5 Model Formulation ................................................................ .............................................. 86 5. I Introduction ................................................................................................................................ 86 Vil

5 .2 Mass Transport in Anode and Cathode ...................................................................................... 89 5.2.1 Reaction Rate Calculation ................................................................................................... 92 5 .3 Charge Transport ....................................................................................................................... 95 Chapter 6 SOFC Modelling, Calibration and Validation ..................................................................... 99 6.1 H2/H20 Model Validation and Discussion ............................................................................... 100 6.2 CO/C02 Model Validation and Discussion .............................................................................. 105 6.3 Comparison between H2 and CO Electrochemical Oxidation ................................................. 109 6.4 Syngas Model Validation and Discussion ............................................................................... 110 Chapter 7 Conclusions and Recommendations .................................................................................. 121 7.1Conclusions .............................................................................................................................. 121 7.2 Recommendations .................................................................................................................... 125 Appendix A Matlab Codes ................................................................................................................. 127 Appendix B Derivation ofrsl rate expression .................................................................................. 134 Appendix C Arrhenius Plots for Each Limiting Steps ....................................................................... 13 7 Bibliography ...................................................................................................................................... 146

Vlll

List of Figures Figure 2.1: Diagram of a single fuel cell (Lisbona et al., 2005) ............................................................. 7 Figure 2.2: Schematic diagram of SOFC operation ............................................................................. 11 Figure 2.3: SOFC design (Singhal, 2000; Yamamoto, 2000) .............................................................. 19 Figure 2.4: Ideal performance of fuel cells ........................................................................................... 21 Figure 2.5: Actual performance of fuel cells ........................................................................................ 22 Figure 3.1: Procedure for preparing NiO powder ................................................................................. 36 Figure 3.2: Catalysis activity test station .............................................................................................. 37 Figure 3.3: Quartz tube reactor. ............................................................................................................ 39 Figure 3.4: a) Electrolyte-anode bilayer; b) NiO/SDC anode-supported cell ..................................... .43 Figure 3.5: A schematic diagram of SOFC test station ........................................................................45 Figure 3.6: SOFC cell set-up ................................................................................................................ 47 Figure 4.1: Methane conversion for SOC and YSZ at 750 and 650°C (SIC= 3, GHSV - 140 h- 1)

.....

50

Figure 4.2: Methane conversion for Ni-SOC and Ni-YSZ catalysts at different temperatures (SIC = 3, GHSV ~ 140 h- 1) ... . .............. ..... .. .................................. . ................. .. ................................................... 52 Figure 4.3: H2 yield for Ni-YSZ and Ni-SOC at different temperatures (SIC=3; GHSV- 140 h- 1) . .... 54 Figure 4.4: CO yield for Ni-YSZ and Ni-SOC at different temperatures (SIC= 3; GHSV-140 h- 1)

... 54

Figure 4.5: C02 conversion for reverse WGS reaction over SDC and YSZ at different temperatures 1

(H 2/C02 = 1, GHSV ~ 70 h- ) .... .. .... ............ ............. . ............................................... ............................ 56 Figure 4.6: C02 conversion for reverse WGS reaction over SOC and YSZ at different temperatures (H2/C02 = 3, GHSV - 70 h- 1) ...................................... .. ............ .......................................... .. ............... 56 Figure 4.7: C02 conversion for reverse WGS reaction over SDC and YSZ at different temperatures (H2/C02 = 4, GHSV - 70 h- 1) .................................... ......... .. ........... .. ................................................... 57 Figure 4.8: C02 conversion for reverse WGS reaction over SOC at different temperatures and H21C02 ratios (GHSV - 70 h' 1)

.......................... . .. .. ...... ... ............. ... .............................. . ......... .. .. .......... . ...... ... . 58

Figure 4.9: C02 conversion for reverse WGS reaction over YSZ at different temperatures and Hi C0 2 ratios (GHSV ~ 70 h- 1)

............................................ .................................... ............. ......... ........ ...... ... ..

58

Figure 4.10: Reaction rate versus flow rate for different particle diameters at two different temperatures (C02/H2=1, GHSV=640h-1)

.................. .. ....... ........... ..................... ............................. ....

60

Figure 4.11: Comparison ofC02 conversion to the equilibrium conversion at two different temperatures for particle size of 210 micrometer and flow rate of 280 ml/min (COiH2=1, GHSV=640h-1)

.. . ..................... . ...... . ..... . ...... .. .................. . ............ .......... ............ .. ................................ 61

lX

Figure 4. 12: Comparison between experiments and simulation for the conversion of C0 2 at 800°C. 70 Figure 4.13: Comparison between experiments and simulation for the conversion of C0 2 at 750°C. 71 Figure 4.14: Comparison between experiments and simulation for the conversion of C02 at 700°C . 71 Figure 4. 15: Comparison between experiments and simulation for the conversion of C02 at 650°C. 72 Figure 4.16: Arrhenius plot of k ........................................................................................................... 73 Figure 4.1 7: Arrhenius plot of Ks ........................................ ................................................................ 73 Figure 4.18: Arrhenius plot of KH20························ ········································ ····································· 74 Figure 4.19: Comparison between experiments and simulations for the conversion of C0 2 at temperature of 800-650°C .................................................................................................................... 75 Figure 4.20: The gradient plots for parameters k, Ks2, Keo, Kc02 and KH2o as a functi on of the catalyst weight using data points at 1023 K ...................................................................................................... 79 Figure 4.21 :The gradient plots for parameters Keo and Ke02 as a function of the catalyst weight using data points at I 073 K ........................................................................................................................... 80 Figure 4.22: MCMC output values for parameters k, Ks and KH20 at a temperature of 1073 K ...... 81 Figure 4.23 : A 95% joint confidence region for parameters k and KL and Ek and EKL (below) ....... 82 Figure 4.24: A plot of predicted values compared with experimentally observed values ................... 83 Figure 4.25: Residual plot that measures the difference between the observed and the predicted values ····························································································································································· 83 Figure 5.1: Electron conductivity of SSC (Hui et al., 20 I 0) ................................................................ 97 Figure 6.1: Experimental (dotted lines) and simulated (solid lines) cell performance using 3% humidified H2 as fuel source at 700, 650 and 600°C ........................................................................ 101 Figure 6.2: H2 and H 20 molar fractions at 700°C for two different cell voltages (0. 7 and 0.5V) ..... I 02 Figure 6.3 : 0 2 molar fractions at 700°C for two different cell voltages (0.7 and 0.5V) .................... 102 Figure 6.4: H2 and H2 0 mo lar fractions at a cell voltage of 0.5 V for three different temperatures .. 103 Figure 6.5: 0 2 molar fractions at a cell voltage of 0.5 V for three different temperatures ................ I 04 Figure 6.6: local current density profile along the anode thickness for humidified H 2 at 700°C. ..... I 04 Figure 6 .7: Experimental (dotted lines) and simulated (solid lines) cell performance using 20%C0/80%C02 at 700, 650 and 600°C ........................................................................................... 105 Figure 6.8: CO and C0 2 molar fractions at 700°C for two different cell voltages (0.7 and 0.5V) and a feed gas composition of20%C0/80%C02 ....... ................... . ........ . .. .... . ......... . ......................... .......... 106 Figure 6.9: 0 2 molar fraction at 700°C for two different cell voltages (0. 7 and 0.5V) and a feed gas composition of20%C0/80%C0 2 ••• ••.•••••. ••••.•..••••.•.• •••.•.•••• ...••• •• .••• . . ..••••• •.•• ••.•.••••••...••...•.••••.•••••••.. .••• 107

x

Figure 6.10: CO and C02 molar fractions at a cell voltage of 0.5V for three different temperatures and for a feed gas composition of 20%C0/80%C0 2 .... .. .......................................................................... I 08 Figure 6.11: 0 2 molar fractions at a cell voltage of 0.5V for three different temperatures and for a feed gas composition of 20%C0/80%C02 ...... ..... ... ..... ......................................... ............................. 108 Figure 6.12: Polarization curve for 20%H2/80%H20 (shown as "Hi'' in the figure legend) and for 20%C0/80%C02 (shown as "CO" in the figure legend) ................................................................... 109 Figure 6.13: Experimental (dotted lines) and simulated (solid lines) cell performance for syngas from diesel, pre-reformed natural gas and biomass gasification at 700°C .................................................. l l l Figure 6.14: Simulation results without WGS on SOC (dotted lines) and with WGS on SOC (solid lines) at 700°C .................................................................................................................................... 112 Figure 6.15: H2 mole fraction without WGS on SOC (solid lines) and with WGS on SOC (dotted lines) incorporation at 700°C and at two different cell voltages (0.7V and 0.4V). a) biomass syngas, b) diesel syngas, c) pre-reformed natural gas ......................................................................................... 114 Figure 6.16: CO mole fraction without WGS on SOC (solid lines) and with WGS on SOC (dotted lines) incorporation at two different cell voltages (0.7V and 0.4V). a) biomass syngas, b) diesel syn gas, c) pre-reformed natural gas .................................................................................................... 115 Figure 6.17: Reaction quotient without WGS on SOC (solid lines) and with WGS on SOC (dotted lines) at 700°C and for two different cell voltages (0.7V and 0.4V). a) biomass syngas, b) diesel syngas, c) pre-reformed natural gas .................................................................................................... 11 7 Figure 6.18: Reaction rate of WGS on SOC (solid lines) and reaction rate of WGS on N i (dotted lines) at 700°C and at two different cell voltages (0.7V and 0.4V). a) biomass syngas, b) diesel syngas, c) pre-reformed natural gas .................................................................................................... 119

Xl

List of Tables Table 2.1: Types of fuel cells (Li, 2006; O'Hayre et al., 2006) ............................................................. 7 Table 2.2: Anodic and cathodic reactions of fuel cells (Selman and Lin, 1993) ................................... 8 Table 2.3: Summary of various key SOFC model developments ........................................................ 31 Table 3 .1: Methane steam reforming evaluation parameters ............................................................... 41 Table 4.1 : Ratio of C02 conversion over SDC to YSZ ........................................................................ 57 Table 4.2: Kinetic study experimental data ......................................................................................... 61 Table 4 .3: Reverse water-gas shift rate expressions depending on the rate limiting step based on Liu et al. (2010) mechanism ........................................................................................................................... 68 Table 4.4: Kinetic parameters of reverse WGS reaction when the surface reaction I is rate limiting. 72 Table 4.5: Reverse WGS activation energy from literature ................................................................. 76 Table 4.6: The parameters estimates and standard deviation obtained from MCMC analysis using all four temperatures with Tref-= 998 K .................................................................................................... 82 Table 4.7: The parameters estimates and standard deviation obtained from MCMC analysis at 998 K ............................................................................................................................................................. 84 Table 5.1: Constant for the semi-empirical form of flGrxn, To ......................................................... 88 Table 5 .2: Values for cri and ei/k parameters for gases of interest in this work (Reid et al., 1987) ..... 91 Table 5.3: Gas Diffusion Volume (m 3/mol) ......................................................................................... 92 Table 5.4: List of boundary conditions for mass-transport .................................................................. 95 Table 5.5: List of boundary conditions for the ionic transport equations ............................................ 97 Table 5.6: List of boundary conditions for the electronic transport equations .................................... 97 Table 6.1: Simulation parameter at 700°C ........................................................................................... 99 Table 6.2: Species composition of syn gas ......................................................................................... 110

XII

Nomenclature List of English symbols

Binary diffusion coefficient Effective Knudsen diffusion coefficients Effective molecular diffusion coefficient Diffusional driving force acting on species k Activation energy Ea,e

Activation energy for the electronic conductivity

Ea,i

Activation energy for the ionic conductivity

Ecell

Cell potential

g

Reversible fuel cell voltage

F

Faraday's constant

F;

Molar flow rate of species i Flow rate of inert Total flow rate

lo

Exchange current density

J

Current density

k

Reaction rate constant

K

Equilibrium constant

M

Molecular weight

N;

Molar flux of species i

in

Inlet molar flow rate of species i

out

Outlet molar flow rate of species i

ni

n;

p

Pressure

r

Reaction rate

R

Gas constant

T

Temperature

v

Voltage XIII

vi Vref

Molar diffusion volume of species i Relative potential difference between the electronic and ionic conductors

W

Mass of catalyst

X;

Conversion of species i

Yi

Product yield of species i

LJGi

Sc

Standard Gibb's energy of species i Rate of production or consumption of electric charge

List of Greek symbols 7Johm

Ohmic overpotential

7Jact

Activation overpotential

7Jcon

Concentration overpotential

7Jr

Total overpotential

a

Exchange transfer coefficient

eiJ

Characteristic Lennard-Jones energy

p

density in kg/m3

a

Conductivity


Standard electronic conductivity Diameter of the molecular collision Standard ionic conductivity

r

Tortuosity factor for molecular diffusion


Porosity of the porous structure

.00

Collision integral

w;

Mass fraction of species i

List of abbreviations and acronyms

AFC

Alkaline Fuel Cell

CHP

combined heat and power XIV

GHSV

Gas Hourly Space Velocity

IT-SOFC

Intermediate Temperature Solid Oxide Fuel Cell

JCRs

Joint Confidence Regions

LSM

Strontium doped Lanthanum Manganite

MCFC

Molten Carbonate Fuel Cell

MCMC

Markov Chain Monte Carlo

MFCs

Mass-Flow Controllers

MSR

Methane Steam Reforming

Ni-SDC

Nickel-Samaria Doped Ceria

ocv

Open Circuit Voltage

PAFC

Phosphoric Acid Fuel Cell

PEMFC

Polymer Electrolyte Membrane Fuel Cell

RWGS

Reverse Water Gas Shift RWGS

SDC

Samaria-Doped Ceria

SOFC

Solid Oxide Fuel Cell

TPB

Three Phase boundary

WGS

Water Gas Shift

YSZ

Yittria-Stabilized Zirconia

xv

Chapter 1 Introduction One of the major drivers to accelerate the development of fuel cells is the increasing concern about the environmental consequences of the continuous use of fossil fuels for both stationary and transportation applications. With the rising concern about greenhouse gas emissions, many efforts are being pursued to develop more efficient energy conversion devices to replace conventional combustion heat engines. Fuel cell technologies offer efficient and clean conversion of chemical energy of fuels to electrical energy. The waste stream from a fuel cell using H2 fuel contains primarily water and heat, thereby greatly reducing greenhouse gases. Even for the fuel cells that can operate on hydrocarbons, the greenhouse gas emissions can be significantly reduced due to the higher efficiency. Also, the operation of some fuel cells (e.g. solid oxide fuel cell, SOFC) are such that C0 2 capture could potentially be implemented with relatively low penalty since the cell exhaust is composed mainly of C02 and water. Therefore, research focusing on the improvement of performance of SOFCs is increasing.

In addition, since no nitrogen oxides or particulates are emitted, fuel cells are known as a very clean technology. With rising fuel prices and stricter emission control regulations, these capabilities make fuel cells even more attractive.

Solid Oxide Fuel Cell (SOFC) is a type of high temperature fuel cell which operates at about 1000°C and is thus capable of producing both electricity and heat. A very important advantage of SOFC is that it can tolerate many types of fuel, including hydrocarbon fuels, natural gas, synthesis gas (syngas) and humidified hydrogen (H 2) (Shi and Cai, 2006). Furthermore, since SOFC operates at high temperatures, SOFC can reform hydrocarbon fuels internally. Internal reforming in a SOFC simplifies the overall system design because the external reformer can be eliminated. A SOFC system with internal reforming has an inherent advantage in terms of energy efficiency because the heat required for the reforming reaction is supplied by the heat generated by the electrochemical reaction. Moreover, the ceramic I

thickness of the electrodes, and not only at the boundary between the electrode and electrolyte. 1.1 Motivation of the Research

Recently, many efforts have been made toward developing intermediate-temperature SOFC (IT-SOFC) operating in the temperature range 600-800°C, because such temperatures enable the use of low cost metallic interconnects, shorter start-up time and improved long-term stability of cell materials by reducing the material degradation rate. Unfortunately, decreasing the operating temperature increases the ohmic overpotential due to higher ionic diffusion in the electrolyte resistance, thereby reducing electrochemical performance. To alleviate this problem, designs of IT-SOFC aim at reducing the thickness of the electrolyte while increasing that of the anode. In such design, ohmic losses are reduced. Therefore this research focused on anode supported cell which is done via a combination of mathematical modelling and experimental validation. Methane reforming reaction and water-gas shift reaction have been taken into account in the modelling for syngas composition. For the water gas shift reaction, it was assumed that this reaction occurs not only on Ni (like all reported studies assumed), but also on SDC. The electrochemical reactions were assumed to be able to occur along the thickness of the electrodes, and not only at the boundary between the electrode and electrolyte.

There are two specific objectives in this work: 1) Determine the reaction rate expression for water gas shift reaction on SDC. 2) Develop a ID model that can predict the performance of the cell and on the Ni/SDC material under different operating conditions (e.g. temperature, current density, fuel composition).

In order to achieve these objectives, two research scopes have been considered: 1) Kinetic Study: Developed a kinetic expression for the reverse water gas shift reaction combining experiments and calculations using Matlab codes involving a non-linear least square problem (lsqcurvefit command). 3

2) Modelling Study: Developed a one-dimensional mechanistic model of anode supported button cells that considers that electrochemical reaction occurs not only at anode/electrolyte interface to predict SOFC performance. The model was validated with experimental data produced in the SOFC group at the University of Waterloo. 1.2 Research Contributions

1) Developed the first Ni/SDC model validated for vanous fuel compositions; 97%H 2/3%H20, 20%C0/80%C02, syngas compositions (diesel, biomass, prereformed natural gas). At the time of writing this thesis, there is only one paper in the literature (Cui et al., 2010) dealing with SDC-electrolyte SOFC modeling, but their model was validated only with humidified hydrogen. 2) Demonstrated that SDC is active towards the water gas shift reaction (WGS) and determined the kinetic parameters of WGS using Matlab codes involving solving a non-linear least square problem (lsqcurvefit command). The WGS on SDC was then incorporated in the modeling study. 1.3 Thesis Outline

This thesis is organised into seven chapters, as follows: Chapter 1 presents an introduction of the research and discusses the motivation of the research, its objectives and its contribution.

Chapter 2 presents a general discussion about fuel cells and an overview of several studies on SOFC single cell modelling.

Chapter 3 describes the experimental techniques including material preparation to study the activity of methane steam reforming (MSR) and kinetics of water gas shift reaction.

Chapter 4 demonstrates the activity of SDC toward the water-gas shift reaction and describes the kinetic study of the water gas shift reaction on SDC.

4

Chapter 5 describes the model formulation for a ID SOFC model of button cell.

Chapter 6 discusses the results of the modelling study involving H2/H20, CO/C02 and syngas compositions.

Chapter 7 presents the conclusions and recommendations for this work.

5

Chapter 2 Literature Review 2.1 Introduction

This chapter begins with a general discussion about fuel cells, and then focuses on solid oxide fuel cells (SOFC) including SOFC fuels, materials, cell design and performance. In the last section (section 2.4), an overview of several studies on SOFC modelling is provided. 2.2 Fuel Cells in Brief

The fuel cell, an electrochemical energy conversion device which directly converts chemical to electrical energy, is a promising technology for producing electricity cleanly and efficiently. The invention of fuel cells as energy conversion systems began in the middle of the 19th century (Hirschenhofer et al., 1998; Stambouli and Traversa, 2002). Because of its high energy efficiency and being environmentally friendly, fuel cells are considered to be potentially attractive devices to produce electricity.

Fuel cells exist in different types. There are five major types of fuel cells: phosphoric acid fuel cell (PAFC), polymer electrolyte membrane fuel cell (PEMFC), alkaline fuel cell (AFC), molten carbonate fuel cell (MCFC) and solid oxide fuel cell (SOFC). The types of fuel cells differ primarily by the type of electrolyte they employ, charge carrier, operating temperature, material, fuel tolerance and performance characteristics, as listed in Table 2.1.

Although the five types of fuel cells have different characteristics, the basic structure of all fuel cells is similar. The cell consists of two electrodes called anode and cathode separated by an electrolyte and connected to an external circuit. A schematic representation of a fuel cell with the reactant and product and the ion conduction flow directions through the cell is shown in Figure 2.1.

6

Table 2.1: Types of fuel cells (Li, 2006; O'Hayre et al., 2006) Type of fuel cells AFC

Characteristic

PEMFC

PAFC

Electrolyte

Polymer membrane H+

Liquid H3P04 (immobilized)

Operating temperature (°C)

50-80

MCFC

SOFC

Liquid KOH (immobilized)

Molten carbonate

Ceramic

Off

co~-

oL·

160-220

60-220

600-700

600-1000

Catalyst

Platinum (cathode I anode)

Platinum (cathode I anode)

Platinum (cathode I anode)

Cell component

Carbon based

Carbon based

Carbon based

Nickel (cathode I anode) Stainless based

Ni: anode LaSrMn03 : cathode Ceramic based

Fuel compatibility

H 2, methanol

H2

H2

H2, CHi

Fuel efficiency (Chemical to electrical)

45-60

55

40-60

60-65

Charge carrier

Fuel

w

1'

f

Poirtr.-e icrr

a

'

N9Jaiwa

or

Prod~ci

Gase: Oll.

....J

55-65

Oiidan: 1r

'

,,,o

Ht ~

Depleted Fuei 1n:l

co

01

H,

~

H2, CH4,

Eedrc¥te (on C«1Uto~

l..

C~pl etec OKcart cm Prti:ucf 33.: 9; :JLi.

L w.::.il

Anccie _J

Figure 2.1: Diagram of a single fuel cell (Lisbona et al., 2005)

In a typical fuel cell, gaseous fuels (e.g. hydrogen) are fed continuously to the anode and an oxidant (typically oxygen from air) is fed continuously to the cathode. The electrochemical reactions take place in the electrodes to produce an electric current. The function of the electrolyte is to conduct ionic charges between the electrodes (Lisbona et al., 2005). Individual fuel cells have a maximum output voltage on the order of 1 V. Substantial 7

voltages and power outputs are obtained by connecting many cells electrically in series to form a fuel cell stack. All five fuel cell types have different electrochemical reactions. The electrochemical reactions occurring at the anode and cathode sides are summarized in Table 2.2. Table 2.2: Anodic and cathodic reactions of fuel cells (Selman and Lin, 1993) Fuel Cell

Anode Reaction

Cathode Reaction

PEMFC

Hz~

2H+ + ze-

Oz+ 4H+ + 4e- ~ 2H z0

PAFC

Hz~

2H+ + ze-

Oz+ 4H + + 4e- ~ 2Hz0

AFC MCFC

Hz+ 20H-

~

2Hz0 + 2e-

Hz +co§- ~ HzO + COz + 2e-

Oz+ 2Hz0 + 4e-

~

40H-

Oz + ZCOz + 4e- ~ 2co§-

CO+ co§-~ 2COz + zeSOFC

Hz+ o z- ~ HzO + 2e-

Oz + 4e-

~

2oz-

CO + oz- ~ COz + ze-

The operational principles of fuel cells and batteries have similarities: both are galvanic cells. They consist of an anode and a cathode in contact with an electrolyte. Both devices generate electrical energy by converting chemical energy using an electrochemical reaction. These reactions occur at the anode and cathode with the electron transfer forced through an external load in order to complete the reaction. Individual cells of both batteries and fuel cells generate only small voltages, which are then combined in series to achieve substantial voltage and power capacities.

Fuel cells differ from batteries in which the chemical reactants are stored. In a battery, the anode and cathode are consumed during use. Thus, a battery can only operate until these materials are fully consumed after which it either must be replaced or recharged, depending on the nature of the materials. In a fuel cell, the chemical reactants are supplied from an external source so that its materials of construction are never consumed and do not need to be recharged. A fuel cell continues to operate as long as reactants are supplied and the reaction products are removed (O'Hayre et al., 2006). 8

The choice of electrochemical device, either battery or fuel cell, depends upon use. For larger scale applications, fuel cells have several advantages over batteries including smaller size, lighter weight, quick refuelling and longer range.

Fuel cells and combustion engines also share some similarities. Both fuel cells and internal combustion engines completely oxidize the fuel. Fuel cells use pure hydrogen or a reformate gas mixture. Internal combustion engines typically use hydrogen containing fossil fuels directly, although they could be configured to operate using pure hydrogen. Both systems use air as the oxidant. In some respects, fuel cells and internal combustion engines are fundamentally different. Fuel cells react the fuel and oxidant electrochemically whereas internal combustion engines react the fuel and oxidant through combustion. Like other electrochemical devices, fuel cells are not limited by the Carnot efficiency as combustion engines are (Lisbona et al., 2005). For exan1ple, when ethanol is burned in a combustion engine, the energy efficiency is limited by the Carnot efficiency and can reach, in practice, only about 25%. This fuel efficiency can be significantly increased when ethanol is first converted to hydrogen and then used in a fuel cell with an efficiency of more than 50% (Rass-Hansen et al. , 2007).

Applications of fuel cells are in transportation, power generation and in powering mobile devices (Shi and Cai, 2006). The application of fuel cells in the transportation sector increases fuel efficiency, decreases foreign oil dependency and becomes an important technology to fight climate change. As fuel cell vehicles begin to operate on fuels from natural gas or gasoline, greenhouse gas emissions will be reduced. The PEM fuel cell is regarded as ideally suited for transportation applications due to its high power density, high energy conversion efficiency, compactness, lightweight nature and low operating temperature (below 100°C).

For stationary power generation applications, both low-temperature and high-temperature fuel cells could be utilized. The low-temperature fuel cells have the advantage that usually a 9

faster start-up time can be achieved, which makes it more attractive for small-power generation. The high-temperature systems such as SOFC and MCFC generate high-grade heat which can be used directly in a heat cycle or indirectly by incorporating the fuel cell system into a combined cycle. SOFC and MCFC are more suitable for large-scale power plants (Dokiya, 2002). SOFCs are expected to play a significant role in residential combined heat and power (CHP) applications (1 to 10 kW) and commercial CHP applications (up to 250 kW), or power plant stationary applications (Wei Zhang, 2006).

2.3 Solid Oxide Fuel Cell SOFC is a high temperature fuel cell operating around 1000°C for state-of-the-art SOFCs, composed of a YSZ (Yittria-Stabilized Zirconia, (Y203)0.os (Zr02)0.92) electrolyte, Ni-YSZ anode, LSM (Strontium doped Lanthanum Manganite, La1-xSrxMn03) and cathode (Singhal, 2000; Zhu and Deevi, 2003). An advantage of the high operating temperature is the possibility of combined heat and power generation which would even further increase the efficiency of this technology (Singhal, 2000). An even more important advantage of SOFC compared to low temperature fuel cells (e.g. PEMFC) is not only the lower cost of the electrocatalyst (Ni for SOFC, as opposed to Pt for PEMFC), but also that they are tolerant to CO, making SOFC fuel flexible (Stambouli and Traversa, 2002). Moreover, the high operating temperature allows internal reforming of the fuel to form H2 and CO, where the heat released by the electrochemical reaction can be utilized by the endothermic steam reforming reaction (Ahmed and Foger, 2000). Internal reforming can also lower the overall system costs because steam required for the steam reforming can be obtained from the steam generated by the electrochemical fuel cell reaction, and because of reduced maintenance due to the elimination of an external reformer (Clarke et al., 1997; Boder and Dittmeyer, 2006; Cheekatamarla et al., 2008). All of these advantages make the SOFC an even more attractive means for producing electrical power.

10

z

1/2 0 2 +2e- --t0 2-

Cathode

t o'- l 0'-r~

Electrolyte

co+o 2---tco 2 +2e- : _ / \ / \ / \ _

Anode

Fuel (H2, CO)

H2 +0 2---tH 2 0+2e-

d

Figure 2.2: Schematic diagram of SOFC operation

Figure 2.2 shows the main components of a SOFC, consisting of two electrodes, called anode and cathode, separated by a dense solid electrolyte. The electrodes are porous to facilitate the transport of fuel and oxidant from the gas channels to the three phase boundaries where electrochemical reactions occur. Within an SOFC anode structure, the hydrocarbon's fuel may be reformed via heterogeneous reaction to produce Hz and CO in the presence of a reforming catalyst (e.g. Ni). The CO may further react with HzO to form Hz and COz via the water- gas-shift reaction. Within the anode, the pore spaces are typically sufficiently- small that the most likely collisions are between gas molecules and surfaces, and there is very little probability for gas- gas collisions. Consequently, gas phase homogeneous kinetics are usually negligible (Hecht et al. , 2005; Zhu and Kee, 2008).

The electrolyte is dense to keep the gases separated and to allow an oxygen concentration difference between the anode and the cathode. Oxygen ions are produced at the three phase boundaries near the cathode/electrolyte interface and are transported by a solid-state migration mechanism through the electrolyte to the anode/electrolyte interface, where oxygen ions react with the fuel (Badwal, and Foger, 1996). Products generated from the reaction are transported back to the fuel channel through pores. The electrochemical reactions at the anode are: 11

+ 0 2CO + 0 2-

H2

+ zeC02 + ze-

~ H20 ~

(2.1)

The electrochemical reaction at the cathode is: (2.2) The 0 2 - ion is drawn through the electrolyte from the cathode to the anode, while electrons are forced through an external circuit from the anode to the cathode. These electrochemical reactions occur continuously as long as enough fuel and oxidant are supplied to the SOFC.

SOFC Fuel:

As was previously stated, one of the major advantages of SOFC is fuel flexibility: fuels that can be used in a SOFC can be hydrogen (H2 ), carbon monoxide (CO), methane (CH4 ) or some higher hydrocarbons and synthesis gas from solid fuels (coal and biomass) (Li et al., 2010). This feature reduces considerably the cost intensive efforts for producing high quality pure hydrogen, as demanded by other types of low temperature fuel cells.

Hydrocarbon based fuels, such as methane, can be reformed to produce H 2 and CO. Reforming is a chemical process that reacts hydrogen-containing fuels in the presence of steam, oxygen, or both, into a hydrogen-rich gas stream. The resulting hydrogen-rich gas mixture is called reformate. Reforming can be further subdivided according to whether (1) it occurs in a chemical reactor outside the fuel cell (external reforming) or (2) it occurs at the catalyst surface inside the fuel cell itself (internal reforming) , the latter being possible in high temperature fuel cells.

Methane steam reforming is one of the most widely used processes for the production of H 2 and CO mixtures. In fact, methane steam reforming accounts for 95% of the hydrogen produced in the United States (Blaylock et al. , 2009). Methane steam reforming is an endothermic reaction and is normally carried out at temperatures around 700-800°C in the presence of a suitable catalyst. Nickel used as anode material can act as a methane steam 12

reforming catalyst. In the reforming of the methane with steam, the dominant reactions are the following two reactions: CH 4 +H 2 0HC0+3H 2 (steam reforming)

(2.3)

CO+H 2 0HC0 2 +H 2 (water gas shift reaction)

The steam reforming reaction is a slow and highly endothermic reaction, and the water gas shift (WGS) reaction is a fast and weakly exothermic reaction. Therefore, the overall reaction resulting from the methane steam reforming reaction and the water gas shift reaction is highly endothermic. Several authors have assumed that the water gas shift reaction is at equilibrium at the reforming temperature (Nagata et al., 2001).

Methane steam reforming is affected by operating pressure, temperature and the ratio of steam to carbon in the feed gas. Methane steam reforming is favourable at low pressure, high temperature and high steam-to-carbon ratio.

Although internal reforming offers an advantage in terms of reducing the overall system cost, it also poses the problem of carbon deposition with the use of Ni-based catalyst which deteriorates the performance of the cell. Methane tends to dissociate on the surface of the nickel particles, depositing carbon, and the CO produced through MSR can also contribute to carbon deposition, as indicated by the following reactions: CH 4

~

C + 2H 2 (methane cracking)

(2.4)

2CO H C + C0 2 (Boudouard) CO + H2 H C + H2 0 (CO hydrogenation)

The formation of carbon is a serious problem in solid oxide fuel cells fed with hydrocarbons. Although the ability to utilize hydrocarbons as a fuel is an important attribute of SOFC, because of carbon formation problems associated with pure hydrocarbon fuels, practical SOFC systems usually operate with mixtures of H2 , CO, and hydrocarbons (Hecht et al., 2005). In addition to reducing carbon formation, mixing the hydrocarbon fuel with H2 and

CO can also avoid thermal stresses in the SOFC, as regions where the highly endothermic 13