application of rare-earth doped ceria and natural minerals ...1347923/fulltext01.pdf · Även om...
TRANSCRIPT
Application of Rare-Earth Doped Ceria and
Natural Minerals for Solid Oxide Fuel Cells
Yanyan Liu
Doctoral Thesis
Stockholm 2019
Division of Heat and Power Technology
Department of Energy Technology
School of Industrial Engineering & Management
KTH Royal Institute of Technology, Sweden
TRITA-ITM-AVL 2019:24
ISBN: 978-91-7873-277-7
Akademisk avhandling
Som med tillstånd av Kungl Tekniska Högskolan framlägges till offentlig
granskning fär avläggande av teknologie doktorsexamen I fysik fredagen den
27 September 2019 kl. 10:00 I K1, Teknikringen 56, KTH, Stockholm.
© Yanyan Liu, August 2019
Tryck: Universitetsservice US-AB, Stockholm
To my husband
Doctoral thesis / Yanyan Liu
i
Abstract
Although solid oxide fuel cell (SOFC) technology exhibits considerable advantages
as compared to other energy conversion devices, e.g. high efficiency, low emission
and fuel flexibility, its high operating temperature leads to rapid component
degradation and has thus hampered commercialization. In recent years, intensive
research interests have been devoted to lowering the operating temperature from
the elevated temperature region (800-1,000 ) to intermediate or low-temperature
range (<800 ). To achieve this goal, material selection plays a dominant role,
involving improving the conductivity of existing electrolytes and developing new
exploitable materials. This dissertation is focused on enhancing the ionic
conductivity of rare-earth oxides (principally doped ceria) and exploring new
candidate materials (e.g. natural minerals) for low temperature (LT) SOFCs.
In this work, the scientific contributions can be divided into four aspects:
i) To develop desirable superionic conductors, Sm3+/Pr3+/Nd3+ triple-
doped ceria is designed to realize the desired doping for Sm3+ in bulk
and Pr3+/Nd3+ at surface domains via a two-step wet chemical co-
precipitation method. It exhibits high ionic conductivity, 0.125 S cm-1
at 600 . The SOFC device using this material as electrolyte displays
a high output power density of 710 mW cm-2 at 550 .
ii) To further clarify the individual effect of Pr3+ in the doped ceria, a
single-element (Pr3+) doped ceria is studied, exhibiting a mixed
electronic/ionic conduction property, capable of being employed as the
core component of electrolyte-layer free solid oxide fuel cells (EFFCs).
iii) To investigate various rare-earth doped-ceria materials in double- and
triple-element doping solutions for LT-SOFCs, Sm3+/Ca2+ co-doped
ceria and La3+/Pr3+/Nd3+ triple-doped ceria are synthesized and then
further incorporated with semiconductors, e.g. La0.6Sr0.4Co0.2Fe0.8O3-δ
(LSCF) or Ni0.8Co0.15Al0.05Li-oxide (NCAL), to serve as a
semiconducting-ionic conducting membrane in EFFCs.
iv) To exploit the feasibility of natural mineral cuprospinel (CuFe2O4) as
an alternative material for LT-SOFCs, three different types of fuel cell
devices are fabricated and tested. The device using CuFe2O4 as cathode
exhibits a maximum power density of 180 mW cm-2 with an open
circuit voltage of 1.07 V at 550 °C, while the device using a
Doctoral thesis / Yanyan Liu
ii
homogeneous mixture membrane of CuFe2O4, Li2O-ZnO-Sm0.2Ce0.8O2
(LZSDC), and LiNi0.8Co0.15Al0.05O2 (NCAL) demonstrates an
improved power output, 587 mW cm-2 under the same measurement
conditions.
Based on this work, a new triple-doping strategy is exploited to improve the ionic
conductivity of doped ceria materials by surface- and bulk-doping methodology.
Furthermore, the material developments of single-phase mixed electronic/ionic
conducting doped ceria and doped ceria/semiconductor composites are realized and
verify the feasibility of EFFC technology. Investigations on CuFe2O4 indicate the
utility of natural minerals in developing cost-effective materials for LT-SOFCs.
Keywords:
Low-temperature solid oxide fuel cells; Doped ceria; Material characterizations;
Electrochemical performances; Natural minerals.
Doctoral thesis / Yanyan Liu
iii
Sammanfattning
Även om fastoxid bränslecellers (SOFC) uppvisar signifikanta fördelar jämfört
med andra energiomvandlingstekniker, t.ex. hög verkningsgrad, låga emissioner
och bränsleflexibilitet, leder dess höga driftstemperatur till snabb
komponentdegradering, vilken har hindrat kommersialiseringen. Under de senaste
åren har intensiv forskning ägnats åt att sänka driftstemperaturerna från de höga
temperaturregionerna (800-1,000 °C) till mellanliggande eller låga
temperaturintervaller (<800 ). För att uppnå detta mål spelar materialvalet en
dominerande roll, vilket bland annat innebär att man förbättrar ledningsförmågan
hos befintliga elektrolyter och utvecklar nya material. Denna avhandling fokuserar
på att förbättra den jonledande förmågan hos oxider av sällsynta jordartsmetaller,
huvudsakligen dopad ceriumoxid, samt forskning på nya kandidatmaterial, t.ex.
naturliga mineraler.
I det här arbetet kan det vetenskapliga bidraget delas in i fyra aspekter:
i) Att utveckla en trippel-dopingmetodik för att syntetisera önskvärda
superjoniska ledningsförmågor i Sm3+/Pr3+/Nd3+ dopad ceriumoxid.
Detta material konstruerades med hjälp av en tvåstegs våtkemisk
samutfällningsmetod för att åstadkomma en önskad dopning för Sm3+
i bulk och Pr3+/Nd3+ vid ytdomäner. Materialet uppvisar en hög jonisk
ledningsförmåga, 0.125 S cm-1 vid 600 . En SOFC-enhet som
använder denna trippeldopade ceriumoxid som elektrolyt har uppvisat
en hög effekttäthet på 710 mW cm-2 vid 550 ;
ii) För att ytterligare klargöra den individuella effekten av Pr3+ i det
dopade ceriummaterialet studerades enfas Pr-dopade ceria, vilken
uppvisade en blandad elektronisk/jonisk ledningsegenskap som skulle
användas som kärnkomponent i avancerad elektrolytskiktsfri fastoxid
bränsleceller (EFFC).
iii) Att undersöka olika sällsynta jordartade dopade ceriummaterial i
lösningar med dubbel- och trippelelement (Sm3+/Ca2+ och
La3+/Pr3+/Nd3+) applicerade för SOFC-teknik med låg temperatur. De
dubbel- och tripeldopade ceriummaterialen var sammansatta med
halvledare, dvs Ni0.8Co0.15Al0.05Li-oxid (NCAL) och
La0.6Sr0.4Co0.2Fe0.8O3-δ (LSCF) för att fungera som en halvledande
jonisk ledande membran i EFFCs.
Doctoral thesis / Yanyan Liu
iv
iv) Att utnyttja naturligt kopparspinell (CuFe2O4) som ett alternativt
material för SOFC. För första gången tillverkades tre olika typer av
anordningar för att undersöka den optimala appliceringen av CuFe2O4
i SOFC. Enheten med CuFe2O4 som katodkatalysator uppvisade en
maximal effekttäthet av 180 mW cm-2 med en öppen kretsspänning
1.07 V vid 550 . En effekttäthet på 587 mW cm-2 emellertid
uppnådes från anordningen bestående av ett homogentblandat
membran med CuFe2O4, Li2O-ZnO-Sm0.2Ce0.8O2 och
LiNi0.8Co0.15Al0.05O2.
Baserat på detta arbete utnyttjades en ny strategi för att förbättra
jonledningsförmågan hos dopade ceriummaterial genom yt- och
bulkdopningsmetodik. Vidare verifierades utvecklingen av EFFC-teknikens
tillförlitlighet av enfasad, blandad elektronisk/jonledande dopade ceriumoxid samt
jonledande multidopade ceria-och halvledarkompositer. Dessa resultat visar att
naturliga mineraler kan spela en viktig roll för att utveckla kostnadseffektiva
material för bränsleceller.
Nyckelord:
Lågtemperatur fastoxidbränsleceller; Dopade ceriumoxid;
Materialkarakteriseringar; Elektrokemiska prestanda; Naturliga mineraler.
Doctoral thesis / Yanyan Liu
v
Acknowledgments
Doctoral studying experience is a life-enriching and life-changing journey. Over
the past four-year traveling through the ocean of knowledge, I finally got chances
to appreciate the beauty of discovering and understanding of the unknown, even if
lost and struggling in the mist often happened to me at the beginning. It would never
happen without the support from people around me. Here, I would like to express
my sincerest gratitude to all who gave me support and help during my Ph.D. study.
First, I would like to express my sincere gratitude to my principal supervisor
Professor Andrew Martin for your invaluable support, guidance, and supervision
throughout my doctoral study. You are always generous to your students with your
advice, time and encouragement whenever I encountered problems. It is my honor
to be your student and I really enjoy learning period under your supervision. And I
also would like to acknowledge the financial support from the Chinese Scholarship
Council (CSC) and Division of Heat and Power Technology (HPT), KTH Royal
Institute of Technology for my Ph.D. study and life in Stockholm, Sweden.
I would like to thank my former principal supervisor Professor Bin Zhu for giving
me the opportunity to work on this fascinating project. His inspiring speeches and
significant guidance brought me to get to know the frontier of the field from the
beginning to the end of my Ph.D. study. I also would like to thank my co-supervisor
Dr. Wujun Wang for his heart-warming encouragement and constructive advice for
my research work, especially the guidance to my dissertation writing. My gratitude
goes to Prof. Lyuba Belova, another co-supervisor, for her guidance in materials
preparation and unselfish advice for my doctoral study. In addition, special thanks
are given to my master supervisor Associate Professor Yongfu Tang, who brought
me to this fascinating road and taught me how to become a qualified researcher
when I started my research life.
Along with my study, I have received much help from my colleagues in Fuel Cell
group: Chen Xia, Muhammad Afzal, and Baoyuan Wang for discussing our
research field and endless friendship. I really enjoyed the friendly atmosphere in
our office. I am also grateful to the HPT laboratory staffs for maintaining an
efficient public lab with an open and friendly atmosphere. Furthermore, heartfelt
gratitude to Professor Muhammet Toprak for serving as the opponent at my final
seminar and advance reviewer of my dissertation. I really appreciate your valuable
time and insightful suggestion to improve my thesis. Also, many thanks to Justin
Chiu, Mahrokh Samavati and Anders Malmquist for your crucial feedback and
Doctoral thesis / Yanyan Liu
vi
suggestion for improving this dissertation, especially to Anders for polishing the
Swedish abstract.
Special thanks go to Professor Yongdan Li from Aalto University for your
enthusiastic advice on the direction of my future research. I really appreciate your
invaluable discussions and guidance about the solid oxide fuel cells and other
energy technologies. Besides, I would like to thank Professor Peter Lund also from
Aalto University for your unselfish help to my research work. I still remember your
warm smile and encouraging words when I gave my first oral presentation at an
international academic conference.
I also want to thank all the supportive group members, former and present. Here,
to mention few: Liangdong Fan from Shenzhen University, Yan Wu from China
University of Geosciences, Yixiao Cai from Uppsala University, Wei Zhang and
Lili Wei from Hubei University. I would also like to thank the other colleagues at
the Department of Energy Technology: Yang Gao, Simeng Tian, Tianhao Xu,
Wenshuai Miao, Yang Song, Saman Nimali Gunasekara, Göran Arntyr, Leif
Pettersson, Jens Fridh, Anneli Ylitalo-Qvarfordt, Tianrui Sun, Chang Su,
Chamindie Senaratne, Jeevan Jayasuriya, Tony Chapman and other friends in the
department for creating a friendly working environment. I also want to thank the
rest of my friends in Sweden: Xinlei Yi, Xiaolin Jiang, Ju Shu, Sheng Jiang, Filippa
Modigh, Cynthia Ho Sin Tian, Chao Zheng, Jingru Wu, Chunlei Wang, Ju Li,
Zongwei Ji, Zongfang Wu, and Yu Yang for the nice time and fun in and outside
KTH. My Ph.D. life would not have been so enjoyable without you all.
Sincere thanks to my beloved husband Shaozheng Ji for your unconditional love
and support for my life, as well as honest criticism, constructive suggestion and
inspiring discussions about my research. Thank you for surviving together with me
through the doctoral years. Your enthusiasm and kindness light up my heart
especially when I felt depressed in the difficult time. I have become a better person
and researcher because of you. Finally, I want to thank my beloved family: my
parents, my brother, my sister and sister-in-law, my aunt and uncle for their love
and support all along, as well as my little nephew for giving me fun during video-
time.
Yanyan Liu
August 2019
Stockholm, Sweden
Doctoral thesis / Yanyan Liu
vii
List of Appended Publications
Paper I
Liu, Y., Fan, L., Cai, Y., Zhang, W., Wang, B. and Zhu, B. Superionic conductivity
of Sm3+, Pr3+, and Nd3+ triple-doped ceria through bulk and surface two-step doping
approach. ACS applied materials & interfaces. 9.28 (2017): 23614-23623.
Paper II
Liu, Y., Wei, L., Dong, W. and Zhu, B. A single-phase mixed-conductive Pr-Doped
CeO2 membrane for advanced fuel-to-electricity technology. (Manuscript under
preparation)
Paper III
Liu, Y., Meng, Y., Zhang, W., Wang, B., Afzal, M., Xia, C. and Zhu, B. Industrial
grade rare-earth triple-doped ceria applied for advanced low-temperature
electrolyte layer-free fuel cells. International Journal of Hydrogen Energy. 42.34
(2017): 22273-22279.
Paper IV
Liu, Y., Wu, Y., Zhang, W., Zhang, J., Wang, B., Xia, C., Afzal, M., Li, J., Singh,
M. and Zhu, B. Natural CuFe2O4 mineral for solid oxide fuel cells. International
Journal of Hydrogen Energy. 42.27 (2017): 17514-17521.
Paper V
Wang, B., Wang, Y., Fan, L., Cai, Y., Xia, C., Liu, Y., Raza R., Aken P., Wang H.
and Zhu B. Preparation and characterization of Sm and Ca co-doped ceria-
La0.6Sr0.4Co0.2Fe0.8O3-δ semiconductor-ionic composites for electrolyte-layer-free
fuel cells. Journal of Materials Chemistry A, 4.40 (2016): 15426-15436.
Paper VI
Wang, B., Cai, Y., Xia, C., Liu, Y., Muhammad, A., Wang, H. and Zhu, B.
CoFeZrAl-oxide based composite for advanced solid oxide fuel cells.
Electrochemistry Communications, 73 (2016): 15-19.
Contributions of the author to the publications
In the listed papers I-IV, Y. Liu is the first author and her contributions include:
proposing the idea, performing the literature survey, designing the experiments,
Doctoral thesis / Yanyan Liu
viii
synthesizing the experimental samples, conducting material characterization work,
electrical property and fuel cell performance measurements, processing and
analyzing all the experimental data, as well writing and revising the manuscripts.
The other authors joined partial characterization measurements and acted as the
reviewers of the manuscripts. In paper V and VI, Y. Liu was involved in partial
contributions including synthesizing the experimental materials, performing the
electrical properties measurements, and conducting the fuel cell performance
measurements.
Doctoral thesis / Yanyan Liu
ix
Table of Contents
Abstract .................................................................................................................. ii
Sammanfattning ................................................................................................... iii
Acknowledgements ................................................................................................ v
List of Appended Publications ........................................................................... vii
List of Figures ...................................................................................................... xii
List of Tables ....................................................................................................... xv
Abbreviations and Nomenclature ..................................................................... xvi
Chapter 1 Introduction ......................................................................................... 1
1.1 Challenges of solid oxide fuel cell technology ............................................. 3
1.2 Materials selection for solid oxide fuel cell technology ............................... 5
1.3 Motivations and objectives ........................................................................... 7
1.4 Structure of this thesis ................................................................................... 8
Chapter 2 Solid oxide fuel cell principles and literature review ....................... 9
2.1 Electrochemistry and thermodynamics ......................................................... 9
2.1.1 Open circuit voltage ............................................................................. 10
2.1.2 Efficiency ............................................................................................. 12
2.2 Materials development ................................................................................ 13
2.2.1 Anode materials development .............................................................. 14
2.2.2 Cathode materials development ........................................................... 14
2.2.3 Electrolyte materials development ....................................................... 16
2.3 Nanocomposites for EFFC technology ....................................................... 18
Chapter 3 Experimental methods and techniques ........................................... 21
3.1 Raw materials .............................................................................................. 21
3.2 Sample preparation ..................................................................................... 21
3.2.1 SDC preparation by co-precipitation approach .................................... 21
Doctoral thesis / Yanyan Liu
x
3.2.2 Pr-doped ceria preparation by hydrothermal method .......................... 22
3.2.3 Sm3+, Pr3+ and Nd3+ triple-doped ceria by two-step doping method .... 22
3.2.4 Double- and triple-doped ceria preparation ......................................... 22
3.2.5 Doped ceria/semiconductor composite preparation ............................. 23
3.2.6 LZSDC nanocomposite preparation .................................................... 23
3.2.7 Natural minutral treatment ................................................................... 23
3.3 Fuel cell device fabrication ......................................................................... 23
3.3.1 Conventional SOFC device ................................................................. 24
3.3.2 EFFC device ........................................................................................ 24
3.4 Material characterizations ........................................................................... 25
3.4.1 X-ray diffraction (XRD) analysis ........................................................ 25
3.4.2 Micro-structure analysis ...................................................................... 25
3.4.3 X-ray photoelectron spectroscopy (XPS) analysis .............................. 26
3.4.4 Raman spectra analysis ........................................................................ 26
3.5 Electrical properties .................................................................................... 26
3.5.1 Electrochemical impedance spectroscopy (EIS) analysis .................... 26
3.5.2 Direct-current (DC) polarization analysis ........................................... 26
3.6 Fuel cell performance measurements ......................................................... 27
Chapter 4 Superionic conductivity of triple-doped ceria for SOFC .............. 29
4.1 Material characterizations ........................................................................... 29
4.1.1 Phase and microstructure properties .................................................... 29
4.1.2 Ultraviolet-visible (UV-vis) Raman scattering analysis ...................... 30
4.1.3 XPS analysis ........................................................................................ 33
4.2 Electrical properties .................................................................................... 35
4.3 Fuel cell performance ................................................................................. 37
4.4 Chapter summary ........................................................................................ 40
Chapter 5 Multi-component rare-earth doped ceria for EFFC ..................... 43
5.1 Single-phase Pr-doped ceria for EFFC ....................................................... 43
5.1.1 Structural and morphology characteristics .......................................... 43
Doctoral thesis / Yanyan Liu
xi
5.1.2 XPS analysis ........................................................................................ 44
5.1.3 Raman scattering analysis .................................................................... 47
5.1.4 Electrical conductivity ......................................................................... 48
5.1.5 Fuel cell performance analysis ............................................................. 50
5.2 Rare-earth triple-doped ceria for EFFC ...................................................... 52
5.3 Double-doped ceria for EFFC ..................................................................... 57
5.4 Fuel cell performance comparsion .............................................................. 58
5.5 Chapter summary ........................................................................................ 59
Chapter 6 Natural minerals for EFFC .............................................................. 61
6.1 Material characterization analysis ............................................................... 61
6.2 Electrochemical performance analysis ........................................................ 63
6.3 Chapter summary ........................................................................................ 67
Chapter 7 Conclusions and future outlook ....................................................... 69
References ............................................................................................................ 73
Doctoral thesis / Yanyan Liu
xii
List of Figures
Figure 1.1 Summary of fuel cell families [16]. ...................................................... 2
Figure 1.2 Ionic conductivity of solid electrolytes for SOFC as a function of
temperature [20]. .................................................................................................... 4
Figure 1.3 Schematic illustrations for a) conventional anode/electrolyte/cathode
three-component SOFC; and b) EFFCs. ................................................................. 5
Figure 2.1 Schematic of SOFC working principle. .............................................. 10
Figure 2.2 Ideal and actual potential vs current response in a fuel cell [11]. ....... 12
Figure 2.3 Schematic diagrams of SOFC stack, individual SOFC device and
functional SOFC components [11]. ...................................................................... 14
Figure 2.4 Crystal structures of the three-type cathodes: a) perovskite; b) layered
perovskite; and c) Ruddlesden-Popper oxides [77]. ............................................. 15
Figure 2.5 Schematic image for fluorite-structured oxide [15]. ........................... 16
Figure 3.1 Schematic illustration of the single fuel cell and measurement setup. 24
Figure 3.2 Schematic diagram of fuel cell performance testing setup. ................. 28
Figure 4.1 a) X-ray diffraction patterns of SDC and PNSDC; b) SEM and TEM
micrographs for PNSDC powder. ......................................................................... 30
Figure 4.2 SEM image of PNSDC and corresponding EDS element mapping. ... 31
Figure 4.3 a) UV-vis diffuse reflectance spectroscopy of SDC and PNSDC; UV-
visible Raman spectra of SDC and PNSDC with an excitation laser wavelength b)
λex=532 nm and c) λex=325 nm; d) Schematic description for Raman scattering by
varying the ultraviolet (UV, λex=325 nm) and visible (λex=532 nm) excitation laser
lines to collect the surface and bulk information....................................................32
Figure 4.4 a) XPS survey spectra of SDC and PNSDC samples; b) XPS O 1s
Spectra of SDC and PNSDC samples; c) XPS Ce 3d Spectra of SDC and PNSDC
samples; d) XPS Pr 3d Spectra of PNSDC sample............................................... 34
Figure 4.5 Electrochemical impedance spectra of SDC and PNSDC measured in air
at a) 400 ; b) 450 ; c) 500 ; d) 550 ; e) 600 and f) the empirical
equivalent circuit to analysis the ionic conductivity of SDC and PNSDC materials
in air. ..................................................................................................................... 36
Figure 4.6 a) Temperature dependence of total conductivity and Arrhenius plots for
SDC and PNSDC. The inset represents area specific resistance (ASR) versus
temperature of SDC and PNSDC; b) current (I) versus time (t) curves obtained at a
temperature range of 450 and 600 at an interval of 50 to determine the
electronic conductivity of PNSDC sample. .......................................................... 37
Figure 4.7 a) Typical I-V and I-P curves of fuel cells based on SDC and PNSDC
electrolyte, respectively, the mixed NCAL and electrolyte materials were used as
the electrode material. The measurements were performed at 550 with H2 and
Doctoral thesis / Yanyan Liu
xiii
air as the fuel and the oxidant, respectively; b) Durability test based on Cell II: Ni-
foam/PNSDC+NCAL (anode)|PNSDC (electrolyte)| PNSDC+NCAL
(cathode)/Ni- foam at a constant discharge current density 80 mA cm-2 operated at
530 .................................................................................................................... 38
Figure 4.8 Electrochemical impedance spectra of the symmetric cell using SDC and
PNSDC samples as electrolyte measured under a) air and b) H2/air conditions at
550 .................................................................................................................... 38
Figure 4.9 Schematic diagrams representing a) the overall O2‐ transfer with the
assistance of reduction conversation of Pr3+/Pr4+ ions in the fuel cell; b) the cathode
reaction process and c) brief routes for the O2‐ surface and bulk transfer and
conduction process in PNSDC electrolyte. ........................................................... 39
Figure 5.1 X-ray diffraction patterns of as-synthesized Pr-free CeO2 and Pr-CeO2
samples. ................................................................................................................. 44
Figure 5.2 Representative SEM micrographs for as-synthesized a) CeO2 and c) Pr-
CeO2 samples; TEM images of b) CeO2 and d) Pr-CeO2 samples suffering
hydrothermal process. ........................................................................................... 45
Figure 5.3 XPS spectroscopy: a) survey spectra of CeO2 and Pr-CeO2 samples; b)
Pr 3d spectra for Pr-CeO2; c) O 1s spectra of CeO2; d) O 1s spectra of Pr-CeO2; e)
Ce 3d spectra of CeO2; and f) Ce 3d spectra of Pr-CeO2. ..................................... 46
Figure 5.4 Raman Spectroscopy of CeO2 and Pr-doped CeO2 samples using an
excitation wavelength of 325 nm. ......................................................................... 47
Figure 5.5 a) Nyquist plots for Pr-CeO2 sample measured at 600 under air
atmosphere; b) Arrhenius plot and corresponding temperature dependence of total
conductivity for Pr-CeO2 sample. ......................................................................... 48
Figure 5.6 I-V-P characteristics of fuel cell devices based on a) the CeO2 and Pr-
CeO2 membrane separately measured at the temperature of 550 ; b) Pr-CeO2
operated at various temperatures of 500, 525, 550 and 600 , respectively. ...... 50
Figure 5.7 Schematic diagrams representing a brief route for the O2- mobility in
fuel cell configuration and detailed O2- transfer with the assistance of redox
Pr3+/Pr4+ couple on the surface and bulk of Pr-CeO2 electrolyte in the vicinity of
oxygen applied side. .............................................................................................. 51
Figure 5.8 SEM micrographs for a) raw material and b) as-prepared LCPN; c) TEM
microstructure and corresponding selected area electron diffraction pattern (SAED,
inset image) for LCPN sample. ............................................................................. 52
Figure 5.9 Typical SEM micrographs of a) raw material and c) LCPN; b) and d)
identified the EDS data with the representative counts of the vertical axis obtained
at the yellow rectangular areas. ............................................................................. 53
Figure 5.10 XRD patterns of LCPN and NCAL samples. .................................... 53
Figure 5.11 EIS patterns of a) the 7LCPN-3NCAL composite at different
temperatures and b) the different compositions of LCPN and NCAL (7LCPN-
Doctoral thesis / Yanyan Liu
xiv
3NCAL, 6LCPN-4NCAL, and 3LCPN-7NCAL) under air/H2 condition at 550 .
All the measurements were performed under air/H2 condition under open circuit
conditions. ............................................................................................................ 55
Figure 5.12 a) The I-V (current density and voltage) and I-P (current density and
peak power density) characteristics of various compositions of LCPN-NCAL
feeding with hydrogen and air at 550 ; b) OCV and maximum power density
values (Pmax) for fuel cells based on LCPN-NCAL in various compositions. ...... 56
Figure 5.13 Schematic illustration (left) and SEM micrograph (right) of EFFC
device based on the 7LCPN-3NCAL composite after the fuel cell performance
measurements. ...................................................................................................... 56
Figure 5.14 a) I-V and I-P characteristics and b) EIS of EFFC devices using the
SCDC/LSCF composites in various weight ratios; c) I-V and I-P performance for
the conventional SOFC device and the device using pure SCDC membrane. ..... 57
Figure 6.1 XRD patterns for a) raw material and b) as-treated mineral samples
sintered at 600, 700 and 900 , respectively.. ..................................................... 60
Figure 6.2 SEM images of a) raw mineral, b) the sample sintered at , and c) EDS
image of raw mineral. ........................................................................................... 62
Figure 6.3 Nyquist plots of a) the samples sintered at 600, 700 and 900 measured
in air atmosphere at 550 ; b) the sample sintered at 900 measured in H2/air at
550 . .................................................................................................................. 64
Figure 6.4 Mechanism for the ionic conductivity of the LZSDC-CuFe2O4H
composite. ............................................................................................................. 64
Figure 6.5 Schematic illustrations for a) type I, b) type II and c) type III. ........... 65
Figure 6.6 a) I-V-P curves for three types of devices measured at 550 ; b)
Durability test of Type III at a constant discharge current of 50 mA cm-2. .......... 66
Figure 6.7 Nyquist plots for a) Type I, b) Type II, and c) Type III operated at fuel
cell condition at 550 . ....................................................................................... 67
Doctoral thesis / Yanyan Liu
xv
List of Tables
Table 5.1 Fuel cell performance comparison based on three types of doped ceria
materials. ............................................................................................................... 58
Table 6.1 The main chemical composition of raw chalcopyrite. .......................... 63
Table 6.2 The fitting results of impedance spectra for three types of devices
measured in H2/air condition at 550 . The unit for resistance is Ω cm2. ........... 66
Doctoral thesis / Yanyan Liu
xvi
Abbreviations and Nomenclature
AFC Alkaline fuel cell
ASR Area specific resistance
CTE Coefficient of thermal expansion
DC Direct current
DMFC Direct methanol fuel cell
e- Electron
Ea Activation energy
EDS Energy dispersive spectrometry
EFFC Electrolyte-layer free fuel cell
EIS Electrochemical impedance spectroscopy
EMF Electromotive force
F Faraday constant
FC Fuel cell
GDC Gadolinium doped ceria
H+ Proton
HOR Hydrogen oxidation reaction
I Current
i Current density
IT Intermediate temperature
k Boltzmann constant
Doctoral thesis / Yanyan Liu
xvii
L Thickness, or inductance
LCPN La/Pr/Nd doped ceria
LSCF LaxSr1-xCoyFe1-yO3, La0.6Sr0.4Co0.2Fe0.8O3-δ
LT Low temperature
LZSDC Li/Zn doped SDC
MCFC Molten carbonate fuel cell
NCAL Ni0.8Co0.15Al0.05LiO2-δ
NEFC Non-electrolyte separator fuel cell
NSDC SDC composited with Na2CO3
OO Oxygen ion in an electrolyte
OCV Open-circuit voltage
ORR Oxygen reduction reaction
P Power density
PAFC Phosphoric acid fuel cell
PEMFC Proton exchange membrane fuel cell
Pi Partial pressure for hydrogen, oxygen and steam
PNSDC Sm3+, Pr3+ and Nd3+-doped ceria
R Resistance
Ro Ohmic resistance
RE Rare-earth
Rp Electrode polarization resistance
S Active area
Doctoral thesis / Yanyan Liu
xviii
SCDC Ca0.04Sm0.16Ce0.8O2-δ
SDC Samarium doped ceria
SEM Scanning electron microscopy
SFM Sr2Fe4/3Mo2/3O6
XPS X-ray photon spectrum
STO Strontium titanate
SIMFC Semiconductor-ion membrane fuel cell
SOFC Solid oxide fuel cell
T Temperature
TEM Transmission electron microscopy
••
OV Oxygen ion vacancy
XRD X-ray diffraction
YSZ Yttria-stabilized zirconia
Z′ Real impedance
Z″ Imaginary impedance
η Overpotential
σ Conductivity
ϕ Phase angle
ω Angular frequency
Doctoral thesis / Yanyan Liu
1
Chapter 1
Introduction
With continuously increasing energy demand and related global warming problems,
the development of clean and high-efficient energy utilization technologies, as
alternates to fossil fuel resources, is an urgent issue [1]. Renewable energy
resources, e.g. solar energy, wind power, hydropower, bioenergy, and wave power,
are vital towards achieving a CO2-neutral energy system and have attracted
considerable attention in recent decades. The development and implementation of
renewables-based technologies have been hindered by a number of issues,
including the intermittency of renewable energy supplies in relation to demand [2-
3]. Hydrogen as an energy carrier has the ability to bridge this gap, provided that
its supply and distribution can be achieved in a cost-effective manner [2,4,5]. A
fuel cell is an integral part of hydrogen-based technologies, enabling the supply of
electricity and potentially heat efficiently from the stored chemical energy of
hydrogen or related fuels [6,7].
Fuel cell technology has a long history since the first fuel cell device (so-called ‘gas
battery’) was demonstrated by Sir William Grove in 1839 [8]. As an
electrochemical energy conversion technology, the fuel cell is not limited by the
Carnot cycle and is capable of supplying electricity with high efficiency in a very
wide range of capacities [9,10]. However, the practical applications of fuel cells
still lag far behind other technologies, for instance, gas turbines or internal
combustion engines. Currently, fuel cell technology is making an appearance in
electrical power markets and for transportation applications [11–15].
A fuel cell device is composed of three major components: electrolyte sandwiched
between two porous electrodes (cathode and anode). In a typical oxygen ion-
conducting fuel cell, the oxygen is reduced primarily at the cathode, and the
resulting oxygen ions (O2-) are transferred through the electrolyte layer approaching
Doctoral thesis / Yanyan Liu
2
to the anode. Subsequently, the O2- ions react with the fuel (e.g. hydrogen or
hydrocarbon) at the anode, with the supply of electrons to an external circuit. A fuel
cell can be classified in different ways depending upon its operating conditions [1].
Based on the types of electrolyte, they are classified as follows (also see Figure
1.1):
i) Proton exchange membrane fuel cells (PEMFCs);
ii) Phosphoric acid fuel cells (PAFCs);
iii) Alkaline fuel cells (AFCs);
iv) Molten carbonate fuel cells (MCFCs);
v) Solid oxide fuel cells (SOFCs).
Figure 1.1 Summary of fuel cell families [16].
Among these fuel cell types, PEMFC, PAFC, and AFC stacks require hydrogen
with relatively high purity to supply the anode, while the elevated-temperature
Doctoral thesis / Yanyan Liu
3
MCFCs and SOFCs can be fed with both hydrogen and carbon monoxide [16].
Comparing to the other fuel cell types, SOFCs have at least four advantages: i) high
current and power density; ii) high fuel utilization factor; iii) available non-noble
metal catalysts; and iv) low cathode/anode polarization losses [17]. The efficiency
of SOFC converting chemical energy to electricity can reach up to 45%~60% with
potential total efficiency of >85% in combined heat and power applications
[11,18,19]. Furthermore, the fuel options for SOFCs are relatively broad, allowing
them to be employed with existing hydrocarbon fuel infrastructure [11,16].
Therefore, SOFC technology is considered to be one of the most promising
candidates currently under research and development for fuel cells.
1.1 Challenges of solid oxide fuel cell technology
The key technical issue for the development of SOFC technology is its high
operating temperature, which can cause rapid system degradation, operational
complexity and high cost [11]. In principle, a SOFC device can be designed for
operating within a wide temperature range (500-1000 ). However, the actual
operating temperature of SOFCs is limited by the ion-conducting capability of the
solid electrolytes at low temperature (see Figure 1.2). A highly ionic conducting
electrolyte is indispensable towards achieving satisfactory fuel cell performance.
Most electrolyte materials, typically (ZrO2)0.9(Y2O3)0.1, require high operating
temperature to achieve desired ionic conductivity (e.g. >0.1 S cm-1) [20]. For
instance, the yttria-stabilized zirconia (YSZ) electrolytes can exhibit good ion-
conducting properties only under high temperature of above 950 [16,21]. Over
the past several decades, considerable efforts have been paid to lowering the
operating temperature of SOFCs to an intermediate temperature range (650-800 )
and even down to a low-temperature range of <650 [11,15,16]. The distinct
benefits of intermediate- or low-temperature SOFCs are the reduced system costs
due to mitigated performance degradation and wider material choices for the
interconnects and seals [22,23]. In addition, the theoretical efficiency of SOFCs is
expected to be enhanced by reducing the operating temperature [11]. Towards this
end, Goodenough [15,24,25] proposed to design oxide-ion conductors, e.g.
La0.8Sr0.2Ga0.8Mg0.2O3 (shown in Figure 1.2), allowing rapid ion transfer at a low
enough temperature level [15,24,25]. Extensive attempts have been devoted to
exploring highly conductive ionic conductors and developing new SOFC
configurations.
Doctoral thesis / Yanyan Liu
4
Figure 1.2 Ionic conductivity of solid electrolytes for SOFC as a function of
temperature [20].
To achieve high ion conductivity, the key issue is to decrease polarization losses,
associated with electrode reaction kinetics and electrolyte conduction at low
temperature (e.g. [16,21,26]). A promising approach in minimizing polarization
losses of LT-SOFCs is to reduce the thickness of electrolytes, transitioning from
electrolyte-supported fuel cells to electrode-supported fuel cells [26–28]. For
example, Barriocanal et al. [12] reported a tremendous ionic conductivity
enhancement (eight orders of magnitude) in YSZ and strontium titanate (STO)
epitaxial heterostructures using ultrathin films of 1~62 nm, showing a potential
application of SOFC materials at slightly above room temperature. Thus, the
Doctoral thesis / Yanyan Liu
5
attainably high ion-conduction and extremely small polarization losses for ultrathin
electrolytes demonstrate a possibility for the SOFCs at low operating temperature.
Designing new fuel cell structures with desired performance is another route to
realize LT-SOFC technology. Recently, a newly developing fuel cell technology,
namely electrolyte-layer free fuel cells (EFFCs), has been given attention. The
schematic illustration of the EFFC device is presented in Figure 1.3. This
technology, with an initial name non-electrolyte separator fuel cell (NEFC), was
first presented in 2011 [30–32]. Since then, many efforts have been undertaken to
understand the underlying working principles and for materials selection [33–41].
Unlike the conventional anode/electrolyte/cathode SOFC configuration, this fuel
cell device is constructed with a homogeneous two-phase nanocomposite, i.e.
semiconductor and ionic conductor, to replace the three components in
conventional devices. Compared to traditional SOFCs, several significant
advantages have been shown from EFFCs: 1) simplified fuel cell structure, leading
to potentially streamlined manufacturing processes; 2) good thermal and chemical
compatibility; 3) and improved fuel cell performance due to the reduction of
electrode polarization losses [31,32,38,39]. Therefore, EFFC energy technology
could provide a possible path to speed up fuel cell commercialization.
Figure 1.3 Schematic illustrations for a) conventional anode/electrolyte/cathode
three-component SOFC; and b) EFFCs.
1.2 Materials selection for solid oxide fuel cell technology
As mentioned previously, a typical single SOFC device is composed of three
components: porous cathode/anode, and a dense electrolyte. For the selection of
electrode materials, three simultaneous purposes must be fulfilled: i) providing the
Doctoral thesis / Yanyan Liu
6
electrochemical reaction sites; ii) providing the pathways for mass transfer (ions
and electrons); iii) collecting electrical current. Hence, the requisite electrode
properties include a porous structure, high electrical conductivity, chemical
stability in reducing/oxidizing operating conditions, and good mechanical
characteristics [1]. Regarding the conventional electrolyte selection, key factors
include: i) sufficiently high ionic conductivity; ii) low electron transport; iii) good
thermodynamic and chemical stability; iv) compatibility with electrodes; v) reliable
mechanical properties [10]. YSZ is the most popular choice for solid electrolytes,
although it requires a relatively high operating temperature to realize satisfactorily
high ionic conductivity [15]. In addition, other fluorite structure oxides, e.g. doped
ceria, have been developed to explore new electrolyte materials for LT-SOFCs
[22,42].
Ceria-based oxides, e.g. typically Ce0.8Sm0.2O1.9 (SDC) and Ce0.8Gd0.2O1.9 (GDC),
have been demonstrated as excellent electrolyte candidates for intermediate- or
low-temperature SOFCs due to their low activation energy, good catalytic activity,
and high ionic conductivity [43]. As reported in the literature, the conductivity of
doped ceria can be enhanced via an appropriate doping approach using low-valence
cations, meanwhile, the materials’ stability also can be improved [43–46]. However,
the existence of redox couple Ce4+/Ce3+ could lead to redox instability and possible
electronic conducting issue, thus resulting in undesirable structural failure and fuel
cell performance degradation [47]. Several studies have focused on the long-term
durability of high-conducting doped ceria materials [22,48–50]. For instance,
Virkar et al. [48] employed an electron blocking layer, e.g. YSZ, to prevent the
electronic leakage current induced by doped ceria. Wachsman et al. [22] reported a
bismuth-oxide-based material as a favorable conductivity electrolyte applied for
LT-SOFCs. Simone et al. [50] investigated δ-Bi2O3 using alternating layers of
Er2O3-stabilized δ-Bi2O3/Gd2O3-doped CeO2 with highly coherent interfaces,
which demonstrated good chemical stability under reducing conditions at high
operating temperature. Aside from these investigations, nanocomposite and
single/double-doping approaches have also attracted numerous researchers’ interest
in improving the ceria-based oxides’ ionic conductivity [51,52]. Banerjee et al. [53]
developed a ceria-based electrolyte by Ca/Sm co-doped route displaying a
superionic conductivity of 0.122 S cm-1 at 700 . These studies indicate hints that
the performance of doped-ceria could be further improved by surface modification
and multi-doping strategies.
Doctoral thesis / Yanyan Liu
7
To be cost-effective, natural minerals, such as CuAlO2, CuFeO2, and CuInO2, have
been intensively investigated for the applications of solar cells and photocatalysis
owing to their prominent catalytic activity [54,55]. Natural hematite was also
explored as a promising electrolyte of SOFCs [37]. Cuprospinel, i.e. CuFe2O4, is
another example illustrating the practical applications of natural minerals in SOFCs
[56]. CuFe2O4 also offers oxidation protection for SOFC metallic interconnects
along with protection against chromium poisoning [58,59].
1.3 Motivations and objectives
As mentioned in the previous section, the main challenge for SOFC technology is
lowering its operating temperature. Clearly, materials selection is a significant
influencing factor in realizing this goal. To this end, the overall aim of this
investigation is to employ new doping strategies for developing and improving
SOFC electrolyte materials. Both ceria-based electrolytes and natural mineral
CuFe2O4 are examined for this purpose in single, double, and triple doped
arrangements. Systematic investigations on material characterizations are carried
out, and electrochemical properties and fuel cell performance are evaluated for
potential LT-SOFC applications.
The detailed objectives to be achieved in this study are specified as follows:
i) Synthesize superionic conductive rare-earth (e.g. Sm3+/Pr3+/Nd3+)
doped ceria via a two-step wet chemical co-precipitation method, and
investigation of ionic conductivity and feasibility for LT-SOFC
technology. (Paper I)
ii) Develop mixed electronic/ionic conducting single-doped ceria (e.g. Pr-
CeO2) for applications in EFFC technology. (Paper II)
iii) Further develop double- and triple-doped ceria to obtain highly ionic
conducting materials and to produce semiconductor composites for
EFFCs. (Paper III, V and VI)
iv) Exploit the natural mineral CuFe2O4 for the use in EFFC devices,
including the determination of optimal fuel cell configurations. (Paper
IV)
Doctoral thesis / Yanyan Liu
8
1.4 Structure of this thesis
Chapter 1 gives some general background of the hydrogen energy economy, fuel
cells, and solid oxide fuel cells. The specific research objectives of this thesis are
also described.
Chapter 2 provides an overview of recent progress on LT-SOFCs, focusing on the
development of electrolyte materials. The basic working principles of SOFCs also
are stated in terms of thermodynamics and kinetics.
Chapter 3 summarizes the materials as well as characterization methods used in this
thesis work to develop and evaluate electrolytes for SOFC applications. Sample
synthesis, material characterization, and electrochemical properties are included.
In Chapter 4, an approach is presented for improving the ionic conductivity of
doped ceria via a two-step wet-chemical co-precipitation method.
In Chapter 5, the characterization investigations of single-, double-, and triple-
doped ceria materials for LT-SOFCs are presented.
Chapter 6 covers the developments of CuFe2O4 as SOFC materials, based on the
heat-treated raw chalcopyrite. Three types of fuel cell configurations using
CuFe2O4 are] investigated to optimize the CuFe2O4 application for LT-SOFCs.
Chapter 7 presents conclusions and suggestions for future work.
Doctoral thesis / Yanyan Liu
9
Chapter 2
Solid oxide fuel cell principles and literature review
In this chapter fundamental characteristics of fuel cell working principles,
particularly aspects related to electrochemistry and thermodynamics, are briefly
described. This provides the framework to understand how fuel cell performance is
influenced by materials properties and operating conditions [60]. The chapter also
includes an overview of key results pertaining to materials development in SOFCs,
with emphasis placed on emerging technologies, e.g. EFFCs.
2.1 Electrochemistry and thermodynamics
A schematic of the typical SOFC operating principle is depicted in Figure 2.1,
taking the O2--conducting electrolyte case as an example. In this case dry air is
supplied to the cathode, and the molecular oxygen from the gas flow reacts with
the electrons from the external circuit, decomposing it into two oxygen ions. (N2
and other gases contained in air are assumed to be inert and are thus neglected.)
The cathode reaction is described by:
1
2𝑂2 + 2𝑒
− → 𝑂2− (2.1)
Subsequently, the generated O2- ions migrate through the O2--conducting
electrolyte to arrive at the anode. Supplied H2 fuel is oxidized and decomposed into
two H+ ions, thereby releasing two electrons. The H+ ions can combine with O2-
ions with the formation of water, and the generated electrons can be then transferred
to an external electric circuit. The anode reaction is written as:
𝐻2+𝑂2− → 𝐻2𝑂 + 2𝑒
− (2.2)
The mass transfer and electric current are realized by a complete circuit in this
SOFC device. The overall electrochemical reaction is stated as:
Doctoral thesis / Yanyan Liu
10
𝐻2 +1
2𝑂2 → 𝐻2𝑂 (2.3)
Figure 2.1 Schematic of SOFC working principle.
2.1.1 Open circuit voltage
When considering a steady-state energy balance of a SOFC device, the enthalpy of
the input reactants including hydrogen and oxygen is equal to the sum of product
enthalpies, the power output, and the net heat dissipated to the surroundings [60].
Normally, the Gibbs free energy concept is employed in lieu of a classical energy
balance analysis, owing to the presence of electrochemical reactions under the
assumed conditions of constant pressure and operating temperature. The total Gibbs
free energy change (𝛥𝐺𝑓) for this device is indicated by the equation:
𝛥𝐺𝑓 = (𝐺𝑓)𝐻2𝑂 − (𝐺𝑓)𝐻2 − (𝐺𝑓)𝑂2 (2.4)
Here the terms (𝐺𝑓)𝐻2𝑂, (𝐺𝑓)𝐻2 , and (𝐺𝑓)𝑂2 refer to the Gibbs free energy of H2O,
H2, and O2, respectively. This reaction process is spontaneous and
thermodynamically favored due to the higher free energy of the reactants than that
of the products [60]. In an ideal reversible system, the yielded electrical energy
Doctoral thesis / Yanyan Liu
11
(𝑛𝐹𝐸) equals to the released Gibbs free energy (𝛥𝐺𝑓). Therefore, the change of free
energy for this SOFC reaction can be written as:
𝛥𝐺𝑓 = −𝑛𝐹𝐸 (2.5)
in which 𝑛 is the moles of electrons derived from equation 2.3, 𝐹 refers to
Faraday’s constant (96500 C g-1) while 𝐸 represents the reversible circuit voltage.
By the transformation equation, the electromotive force (or theoretical open-circuit
voltage, abbreviated as OCV) is calculated using:
𝐸 = −Δ𝐺𝑓
2𝐹 (2.6)
The electromotive force of a fuel cell is described using the Nernst equation:
𝐸 = 𝐸0 +𝑅𝑇
2𝐹ln𝑃𝐻2 ⋅𝑃𝑂2
0.5
𝑃𝐻2𝑂 (2.7)
where 𝐸0 refers to the ‘ideal’ electromotive force under standard conditions, and
𝑃𝐻2, 𝑃𝑂2 and 𝑃𝐻2𝑂, respectively, are the partial pressures of hydrogen in the anode
chamber, oxygen in the cathode chamber along with generated steam.
The cell voltage (𝑉), by definition, is given by the difference between cathode and
anode potential:
𝑉 = 𝐸cathode − 𝐸anode (2.8)
Ideally, the theoretical voltage of a fuel cell in equilibrium is equal to the
electromotive force calculated by the Nernst equation. However, in practical
applications a fuel cell must supply an electric current, resulting in a lower cell
voltage than the theoretical value. Figure 2.2 demonstrates the actual voltage
versus current response in comparison with an ideal model of the fuel cell. The
actual cell potential is lower than the theoretical one due to various irreversible
losses. Generally, the irreversibility is linked to polarization (or over-potential)
losses, attributing primarily to activation, ohmic and concentration polarizations.
Thus, the actual voltage of a fuel cell is written as:
𝑉 = 𝐸 − ∆𝑉act − ∆𝑉ohm − ∆𝑉con (2.9)
The activation polarization (∆𝑉act) originates primarily from the electrochemical
reaction rates, i.e. hydrogen oxidation reaction occurred in anode and oxygen
reduction reaction occurred in the cathode. Both the reaction kinetics and electrode
geometric effects have an impact on activation polarization [26]. The ohmic
polarization (∆𝑉𝑜ℎ𝑚) is dominated by the ionic conductivity capacity of electrolytes,
but the electronic conductivity of electrodes contributes insignificantly to the ohmic
Doctoral thesis / Yanyan Liu
12
resistance. With respect to the concentration polarization (∆𝑉𝑐𝑜𝑛), the limited mass
transport of reactants and products are contributing factors.
Figure 2.2 Ideal and actual potential vs current response in a fuel cell [11].
2.1.2 Efficiency
The overall efficiency relies on the kinetics and thermodynamics of fuel cells [22].
It can be described primarily using three parameters including the
thermodynamic/Gibbs efficiency (𝜂𝑡ℎ), the voltage efficiency (𝜂𝑣) and the fraction
of fuel used or fuel utilization efficiency (𝜂𝑓). The overall efficiency (𝜂) is given
here [16]:
𝜂 = 𝜂𝑡ℎ × 𝜂𝑣 × 𝜂𝑓 (2.10)
The thermodynamic efficiency is described considering the reaction enthalpy (∆H)
and entropy (∆S). Enthalpy refers to the delivered heat in a reaction and entropy is
associated with the change of reaction order. The well-known equation relating
these terms with Gibbs free energy (𝛥𝐺𝑓) is expressed as:
𝛥𝐺𝑓 = 𝛥𝐻 − 𝑇𝛥𝑆 (2.11)
In the SOFC reaction, 𝛥𝐺𝑓 is less than 𝛥𝐻 and reaction heat is delivered from the
conversion of chemical energy stored in the fuel. Therefore, the thermodynamic
efficiency can be described as [60–62]:
𝜂𝑡ℎ =Δ𝐺𝑓
𝛥𝐻= 1 −
𝑇𝛥𝑆
𝛥𝐻 (2.12)
Doctoral thesis / Yanyan Liu
13
The voltage efficiency (𝜂𝑣) is calculated as the ratio between the practical potential
and the theoretical electromotive force:
𝜂𝑣 =V
E=𝐸−∆𝑉act−∆𝑉ohm−∆𝑉con
𝐸 (2.13)
The fuel utilization efficiency (𝜂𝑓) can be expressed as the ratio between the actual
fuel and the total supplied fuel. In a SOFC operation, the generated current density
is i and the input rate of fuel is 𝑣𝑓 (mol sec-1). Hence,
𝜂𝑓 =𝑖/𝑛𝐹
𝑣𝑓 (2.14)
Substituting Equation 2.9 using 2.11, 2.12 and 2.13, the overall efficiency of SOFC
is expressed as:
𝜂 = (1 −𝑇𝛥𝑆
𝛥𝐻) × (
𝐸−∆𝑉act−∆𝑉ohm−∆𝑉con
𝐸) × (
𝑖/𝑛𝐹
𝑣𝑓) (2.15)
The overall fuel cell efficiency is determined by three aspects, as described in
Equation 2.15. The thermodynamic efficiency, without the limitation of the Carnot
Cycle, suggests that low temperature is favored. The voltage efficiency is
principally affected by the properties of fuel cell materials, e.g. the ionic
conductivity of electrolytes, while the catalytic activity of the electrodes contributes
to the fuel utilization. Therefore, lowering the operating temperature and
developing improved electrode/electrolyte materials to reduce cell polarization and
enhance fuel utilization efficiency are the primary methods to improve overall
SOFC efficiency [15,22].
2.2 Materials development
In practical applications, a SOFC always functions in a system. Generally, a SOFC
stack consists of several individual fuel cells, and there are three different stack
configurations that are widely used: tubular, monolithic and planar SOFC stacks
[1]. The schematic diagrams of the SOFC stack, individual SOFC device, and
functional SOFC components are shown in Figure 2.3. As described previously,
the key technical challenge for the development of the SOFC is to reduce its
operating temperature, thus developing appropriate electrode/electrolyte materials
is the principle issue for SOFC technology. In this section, the material
developments for individual SOFC components are discussed in the context of a
literature survey.
Doctoral thesis / Yanyan Liu
14
Figure 2.3 Schematic diagrams of SOFC stack, individual SOFC device and
functional SOFC components [11].
2.2.1 Anode materials development
In the oxygen-conducting SOFC case, the anode functions in multiple roles: i) to
catalyze the electrochemical oxidation reaction of fuels, e.g. H2; ii) to transfer
oxygen ions from the electrolyte/anode interface into the anode structure; iii) to
transport the generated electrons; and iv) to assist the reactant and product transport
at the reaction sites [21,63,64]. Among various anode materials for SOFCs, the
most widely used ones are porous ceramic and metallic composites of nickel cermet
bonded to ion-conducting materials, most often yttria-stabilized zirconia (Ni/YSZ)
[63–67]. Ni/YSZ has several advantages: low-cost, good high-temperature stability
and matching thermal expansion to the most commonly employed electrolytes [68].
In addition, the nickel cermet serves as an excellent electron conductor and good
electrocatalyst for fuels (particularly hydrogen) at relevant operating temperature,
while YSZ acts as a porous mechanical support and provides ionic conduction
pathways [63,65,66]. However, Ni/YSZ anode has some drawbacks, such as
potential reduction/oxidation cycling instability and low-tolerance to impurities
leading to degraded fuel cell performance [63,69–71]. Therefore, current research
efforts are aimed at developing new anode materials, such as ceria-containing
anodes [72], perovskite-type ceramics [73,74], and layered materials [75].
2.2.2 Cathode materials development
In the overall SOFC process, electrochemical reactions occurring at the cathode
region have been given more attention in comparison to anode-related reactions.
This is can be interpreted by the slower reaction kinetics of oxygen reduction at the
Doctoral thesis / Yanyan Liu
15
cathode than that of hydrogen oxidation at another electrode. Therefore, the cathode
is the primary contributor to the overall polarization resistance of fuel cells.
For SOFC cathode materials, good catalytic activity for oxygen reduction reaction
(ORR) and electron transport functionalities are required. Furthermore, the desired
cathode materials should be able to facilitate mass transfer and be matched
thermomechanically and chemically with other components, i.e. electrolyte and
interconnections [1,10,16,76]. In order to work at low operating temperature,
appropriate cathode materials with good low-temperature catalytic activity for
ORR have been in focus for reducing cathode polarization losses and for achieving
long-term stability [77,78]. Poor ORR catalytic activity is still the key obstacle in
developing commercially viable LT-SOFCs (below 600 ) [78]. To improve
cathodic performance, Shao et. al. [76] proposed two strategies: i) developing new
cathode materials to perform good fuel cell performance in the low-temperature
regime; and ii) optimizing the microstructure of the developed cathode materials to
reduce the cathode polarization resistance.
Figure 2.4 Crystal structures of the three-type cathodes: a) perovskite; b) layered
perovskite; and c) Ruddlesden-Popper oxides [77].
As reported in the literature, three main perovskite-related types, including the
cubic-type, layered-type and Ruddlesden-Popper type perovskites, were selected as
Doctoral thesis / Yanyan Liu
16
promising cathode candidates for LT-SOFC applications [77]. The crystal
structures for the three-type cathode materials are displayed in Figure 2.4. As
shown, all of the three-type cathodes have an oxygen 6-fold coordinated transition
metal scaffold with the alkaline-earth or lanthanides ions-located vertices at a cube
structure [77]. Among these potential cathode materials, La1-xSrxCo1-yFeyO3
(LSCF) and La1-xSrxMnO3 (LSM) have demonstrated excellent mixed electronic-
ionic conducting characteristics and good ORR catalytic activity [76–81].
2.2.3 Electrolyte materials development
For the desired electrolyte, several requirements need to be met: i) sufficiently high
ionic conductivity; ii) negligible electronic conduction; iii) thermodynamically and
chemically stabilizable property during fuel cell operating process; iv) reliable
mechanical strength; v) chemical inertness with anode/cathode during processing;
and vi) compatible thermal expansion coefficient to the anode and cathode
materials [10].
Figure 2.5 Schematic image for fluorite-structured oxide [15].
Doctoral thesis / Yanyan Liu
17
In the published literature, four structural groups of O2--conducting oxides were
employed as electrolytes in SOFCs, including fluorite-type, perovskite, intergrowth
perovskite and pyrochlore-type [82–86]. Generally, oxygen ions can be transported
through the oxygen vacancies in oxygen-ion conductors and the oxygen vacancy
concentration can be enhanced by doping. In a typical fluorite-structured zirconia
(ZrO2) case, the formation of oxygen vacancy with yttria as a dopant can be
described using the defect equation in Kröger-Vink notation [10,86]:
𝑌2𝑂3 → 2𝑌𝑧𝑟′ + 3𝑂0
𝑋 + 𝑉0·· (2.16)
Figure 2.5 demonstrates the fluorite-structured oxide with the formed oxygen
vacancies. As observed, the host cations, commonly Zr4+ or Ce4+, can be replaced
by lower valent ions, e.g. Y3+ or Gd3+, to create oxygen vacancies and thus improve
their ionic conductivities.
For proton-conducting SOFCs, the operating mechanism is different from an
oxygen ion-conducting SOFC device. In this case, hydrogen is catalytically
decomposed into protons on the anode side, and thereafter protons migrate to the
cathode region across the electrolyte. Meanwhile, water vapor is formed on the
cathode side. The electrochemical reaction equations in anode, cathode and the
overall equation, respectively, are expressed as follows:
𝐻2 → 2𝐻+ + 2𝑒− (2.17)
1
2𝑂2 +𝐻
+ + 2𝑒− → 𝐻2𝑂 (2.18)
𝐻2 +1
2𝑂2 → 𝐻2𝑂 (2.19)
The migration of protons in an H+-conducting electrolyte is easier to be realized
than the migration of oxygen ions in oxygen-ion conductors, resulting in lower
activation energy. Thus, a high proton conductivity can be expected, making the
intermediate or low-temperature operation of SOFCs feasible [17,87,88]. Most of
the reported proton conductors are perovskite-type oxides, including BaCeO3- and
SrCeO3-based oxides, or Y-doped barium zirconate [17,87,89,90]. Among these
materials, BaCeO3-based perovskites are verified to demonstrate some of the
highest proton conductivities [87,91].
In the mixed proton/oxygen ion-conducting electrolyte cases, nanocomposites have
been shown to make substantial contributions, especially from the viewpoint of
improving ionic conductivity. Two categories of nanocomposites have been
Doctoral thesis / Yanyan Liu
18
extensively employed as electrolytes for SOFCs. The oxide/carbonate
nanocomposites and particularly ceria or doped ceria/carbonate materials, as the
first type, have attracted attention during the past decade owing to the enhancement
of ionic conductivities via carbonate coatings [92–97]. Oxide/oxide
nanocomposites, representing the second category, typically exhibit mixed
proton/oxide ion conductivities and can be classified into three groups: ionic
conductor/ionic conductor (e.g. YSZ/SrTiO3 [98]), ionic conductor/insulator (e.g.
Gd0.2Ce0.8O2/Mn-Co doped Al2O3 [99]), and semiconductor/semiconductor (e.g.
LiFeO2-LiAlO2 composite [100]).
2.3 Nanocomposites for EFFC technology
Beyond the three groups of oxide/oxide nanocomposites mentioned above, a fourth
type – semiconductor/ionic-conductor composites – has been developed
specifically for EFFC technology. This nanocomposite provides enhanced ionic
conduction with the addition of electronic conductivity. Such materials are not
suitable as conventional SOFC electrolytes owing to the internal short-circuiting
issue. In contrast, the newly developed EFFC concept is able to exploit these traits
by semiconductor/ionic-conductors as homogeneous two-phase materials for
embedding anode and cathode within the electrolyte layer. To elaborate further,
semiconductor/ionic conductor nanocomposites consist of two phases: ion-
conducting phase, and semiconductor phase. Together both are capable of
functioning as a single-component fuel cell. There are several requirements for
semiconductor/ionic nanocomposites in EFFCs [31,32,34,40,101–103]:
i) balanced ionic and electronic conduction;
ii) existence of percolating network for electrons/holes and ions;
iii) existence of P-N junctions formed inside the semiconductor-ionic
composite, to prevent the electrons through the single component layer.
As proposed in the literature, the catalytic roles for the oxidation of hydrogen and
reduction of oxygen are realized through the functional semiconductor/ionic
nanocomposite [101,102]. The electrochemical reactions in an EFFC device can be
described as follows [102]:
At H2-contacting side (hydrogen oxidation):
𝐻2 → 2𝐻+ + 2𝑒− (2.20)
At O2-contacting side (oxygen reduction):
Doctoral thesis / Yanyan Liu
19
1
2𝑂2 + 2𝑒
− → 𝑂2− (2.21)
Overall fuel cell reaction:
𝐻2 +1
2𝑂2 → 𝐻2𝑂 (2.22)
A key fundamental issue to be addressed for EFFC is how to realize the separation
of ions and electrons, thus avoiding the internal short-circuiting problem in a
homogenous semiconductor and ionic conductor layer. Up to now, although
considerable experimental results with high OCVs and comparable power output
have been reported, no internal short circuit occurring has been shown for EFFCs
[30–32,35,41]. Some possible principles to explain the charge separation inside
EFFCs have been proposed based on the physical bulk heterojunction [31,32],
Schottky junction [34] and the energy band alignment [36].
Doctoral thesis / Yanyan Liu
20
Doctoral thesis / Yanyan Liu
21
Chapter 3
Experimental methods and techniques
3.1 Raw materials
The main materials used in this investigation are the single, double, or multi rare-
earth elements (e.g. Sm, Pr, Nd, and La) doped-ceria materials. The used raw
materials include Ce(NO3)2·6H2O, Sm(NO3)2·6H2O, Pr(NO3)3·6H2O,
Nd(NO3)3·6H2O, La(NO3)2·6H2O and Na2CO3. The origin chemicals are analytical-
grade level and are obtained from Sinopharm Chemical Reagent Co. Ltd (China).
Several approaches, including co-precipitation method, solid-state reaction, and
hydrothermal method, are employed here to obtain the desired samples.
The natural cuprospinel comes originally from the Tonglvshan zone, Hubei
Province, China, modified by a simple heat-treatment process.
Commercial multiple oxides, i.e. Ni0.8Co0.15Al0.05Li-oxide, abbreviated as NCAL,
are acquired from Tianjin Bamo Sci. & Tech. Joint Stock Ltd (China).
3.2 Sample preparation
3.2.1 SDC preparation by co-precipitation approach
Samarium doped ceria (denoted shortly as SDC) is synthesized by a carbonate co-
precipitation solution. First of all, a designed stoichiometric ratio of 4:1 for
Ce(NO3)2·6H2O and Sm(NO3)2·6H2O are weighed and mixed with deionized water.
In parallel, a solution of sodium carbonate is prepared as the precipitant for the
above metal salt solution. The molar amounts of sodium carbonate are double that
of the total metal ions. Subsequently, the as-prepared mixed solution is stirred
vigorously for 6 h and then allowed to reset at the ambient temperature for another
6 h. Afterward, the product is washed several times in the deionized water to
remove any unreacted carbonate. The final SDC product is obtained after a 12-h
drying process followed by a 4-h calcination process at 800 .
Doctoral thesis / Yanyan Liu
22
3.2.2 Pr-doped ceria preparation by hydrothermal method
Pr-CeO2 sample is designed to synthesize with Pr3+ dopant as 10% in molar ratio
via hydrothermal method followed by a calcination process. The chemicals,
including Ce(NO3)3·6H2O, Pr(NO3)3·6H2O, and Na2CO3, are used in the
preparation process for Pr-CeO2 material. The precursor of Pr-CeO2 is prepared
firstly through a hydrothermal route. In this typical synthesis, the designed molar
amounts of Ce(NO3)3 and Pr(NO3)3 are dissolved in deionized water. Na2CO3
solution with a molar ratio to the total of metal ions (Ce3+ and Pr3+) of 2:1 is added
into the precursor solution as a precipitant. The precipitation-forming reaction is
carried out at 90 under continuous stirring conditions for 2 h. Subsequently, the
obtained precipitation is transferred into polytetrafluoroethylene reactor maintained
in 120 for 4 h. Then the product from hydrothermal process is sintered in a
muffle furnace at 700 for 2 h, followed by washing repeatedly with distilled
water. As a comparison, a bare CeO2 sample is synthesized in the same procedure.
3.2.3 Sm3+, Pr3+ and Nd3+ triple-doped ceria by two-step doping method
Three elements (i.e. Sm3+/Pr3+/Nd3+) combined to form a triple-doped ceria sample
are synthesized by means of a two-step co-precipitation solution for yielding the
desired elemental surface distribution map. First of all, Sm3+ is doped into the bulk
of ceria, i.e. SDC, via the carbonate co-precipitation method as mentioned above.
Praseodymium and neodymium nitrates (9:1 molar ratios) are used as the secondary
doping ions, introduced by a modified solid-state reaction method. The total molar
ratio of the praseodymium and neodymium nitrates are designed for 1.8 mol. % and
are dissolved in the deionized water to obtain transparent precursor solution.
Subsequently, SDC is added to the prepared solution and homogeneous suspension
is obtained. The final product, PNSDC (Sm3+/Pr3+/Nd3+-doped ceria), is obtained
after a 2-h calcination process at 750 followed by coarse grinding.
3.2.4 Double- and triple-doped ceria preparation
Double-doped ceria powder (Ca0.04Sm0.16Ce0.8O2-δ, SCDC) is prepared by means of
a co-precipitation route. First, the stoichiometric amounts of Ca(NO3)2 (4 mol. %),
Sm(NO3)2·6H2O (16 mol. %) and Ce(NO3)3·6H2O (80 mol. %) are dissolved in
deionized water to form 0.5 mol L-1 solution. Na2CO3 is added into the above salt-
solution as the precipitating agent with stirring vigorously. A white precipitate is
observed in the reactant solution. Subsequently, the resultant product is repeatedly
washed using deionized water to remove any residual Na2CO3. The precursor is put
into a drying oven at 120 for around 10 h. The final SCDC sample is obtained
after a 4-h calcination procedure at 800 followed by grinding.
Doctoral thesis / Yanyan Liu
23
The triple-doped ceria samples (i.e. doping with La3+, Pr3+, and Nd3+, denoted as
LCPN) are synthesized using a co-precipitation method. Primarily, the raw rare-
earth nitrates including La, Pr, and Nd elements are dissolved into the deionized
water and Na2CO3 is used as the precipitating agent. Then, multiple filtration and
washing with deionized water are conducted on the resulting solution followed a 2-
h calcination process at 750 to yield the final LCPN sample.
3.2.5 Doped ceria/semiconductor composite preparation
The doped-ceria/semiconductor composites are prepared via a straightforward
solid-state blending method. First of all, doped-ceria is prepared as described above.
Then four groups of the resulting doped-ceria powders are separately mixed with
NCAL or LSCF (La0.6Sr0.4Co0.2Fe0.8O3-δ) in various ratios (20 wt.%, 30 wt.%, 40
wt.%, and 50 wt.% NCAL or LSCF). These mixtures are ground and calcined at
700 for 30 minutes and then ground once again to obtain homogeneous doped-
ceria/NCAL or doped ceria/LSCF composite samples.
3.2.6 LZSDC nanocomposite preparation
LZSDC nanocomposite is synthesized as reported in the literature [104]. Briefly,
identical molar amounts of LiNO3 and Zn(NO3)2 are dissolved and citric acid is
added as the precipitating agent. (The amount of citric acid is three-quarters of the
integral molar amounts of both nitrates.) Subsequently, a continuous stirring
operation is conducted on this reactant under 120 condition to form a sol-gel
with the addition of SDC powder. The final LZSDC is obtained after a 2-h
calcination procedure.
3.2.7 Natural mineral treatment
Cuprospinel is firstly crushed and ground to 200 meshes. Subsequently, the raw
mineral is directly calcined at different temperature (i.e. 600 , 700 , and
900 ) in air atmosphere lasting for 2 h, respectively to obtain different products.
3.3 Fuel cell device fabrication
In this thesis, two kinds of fuel cell devices are fabricated in terms of the
conventional three-component SOFC configuration and the emerging EFFC
configuration. The devices are co-pressed to form a disc device under the 150 MPa
pressure. The diameter of the fuel cell device is 13 mm and the thickness is about
1 mm. The active area for fuel cell reactions is approximately 0.64 cm2.
Doctoral thesis / Yanyan Liu
24
3.3.1 Conventional SOFC device
The single-phase Sm3+, Pr3+ and Nd3+ triple-doped ceria with super-high ionic
conductivity is used as an electrolyte for LT-SOFC. The NCAL combining with
PNSDC sample is employed as both cathode and anode, denoted as Cell I. A
parallel pellet (Cell II) is fabricated using SDC to replace the PNSDC in Cell I.
Commercial nickel foam is attached on both sides of the pellet as the structure
supports. The fuel cell devices can be briefly written as follow:
Cell I: Ni-foam/SDC+NCAL (anode)|SDC (electrolyte)|SDC+NCAL
(cathode)/Ni- foam
Cell II: Ni-foam/PNSDC+NCAL (anode)|PNSDC (electrolyte)| PNSDC+NCAL
(cathode)/Ni- foam
3.3.2 EFFC device
In the procedure of EFFC fabrication, the commercial NCAL is used and attached
to the devices as the supporting structure. Firstly, NCAL is mixed with terpineol to
obtain a NCAL slurry. Subsequently, the NCAL slurry is pasted on the nickel foam.
This structure is denoted as Ni-foam/NCAL.
3.3.2.1 EFFC device based on single-phase Pr-CeO2
Figure 3.1 Schematic illustration of the single fuel cell and measurement setup.
The single symmetric fuel cell is fabricated in a ‘sandwich’ configuration (Ni-
foam/NCAL | Pr-CeO2 | NCAL/Ni-foam) with Pr-CeO2 sample as core membrane
layer and semiconducting NCAL thin layer pasted onto nickel foam attached to
both sides of Pr-CeO2 layer. The same configuration is constructed using CeO2
Doctoral thesis / Yanyan Liu
25
sample instead of Pr-CeO2. The schematic illustration of a single fuel cell and
measurement setup are presented in Figure 3.1.
3.3.2.2 EFFC device based on triple-doped CeO2
The EFFC devices using LCPN sample are fabricated, similar to the above
configuration. The core component of this device is the LCPN/NCAL composite
with Ni-foam/NCAL employed as supporting structure on both sides of this fuel
cell pellet. In parallel, a range of weight ratios for the core LCPN/NCAL composite
are designed to fabricate various EFFC devices: LCPN:NCAL of 7:3, 6:4, and 3:7,
respectively. In addition, the device using pure LCPN as the electrolyte is also
prepared in the same procedure as the comparative case.
3.3.2.3 EFFC device based on natural mineral
To investigate the optimal fuel cell application for this natural mineral, three fuel
cell devices are fabricated in different configurations. In Type I, CuFe2O4 is used
as the cathode, while NCAL and LZSDC are employed as anode and electrolyte,
respectively. In Type II, the CuFe2O4 is mixed with LZSDC to yield a
CuFe2O4/LZSDC composite with a volume ratio of 1:1. Then the CuFe2O4/LZSDC
composite is appended to both sides of the device as a functional layer. For Type
III, CuFe2O4 material is used combing LZSDC and the semiconducting NCAL as
the core component of the EFFC device. The CuFe2O4/LZSDC composite is mixed
with NCAL in an equal volume.
3.4 Materials characterizations
3.4.1 X-ray diffraction (XRD) analysis
The physical characterizations for the materials used in this thesis are analyzed
using X-ray diffraction (XRD, D-max-2500 X-ray diffractometer, Rigaku Corp.,
Japan) with filtered Cu Kα radiation. The data are recorded in a step-scan mode of
the 2theta range of 10-80o with a step size of 0.02o.
3.4.2 Micro-structure analysis
The morphology and microscopy analysis are carried out on a field emission
scanning electron (FE-SEM, Hitachi S-4800). It is also equipped an energy-
dispersive spectrometer (EDS) to analyze the elemental information of samples.
The transmission electron microscopy (TEM) measurement is carried out on the
equipment of TECNAI20 or JEM-2100, JEOL.
Doctoral thesis / Yanyan Liu
26
3.4.3 X-ray photoelectron spectroscopy (XPS) analysis
X-ray photoelectron spectroscopy (XPS) analysis is performed on a Physical
Electronics Quantum 2000, Al Kα X-ray source. It is conducted to investigate the
surface chemical properties of the used samples, such as SDC, PNSDC, and Pr-
CeO2. The information related to the oxidation states of surface elements, as well
as the atom species can be identified. The IGOR software (Wavemetrics, Lake
Oswego, OR) is used here to fit the XPS original measured data. The chemical
analysis also can be performed based on the fitting results. The binding energies
are calibrated using the C 1s peak at 284.6 eV as a reference and quoted with a
precision of ± 0.2 eV.
3.4.4 Raman spectra analysis
Raman spectra analysis is carried out on a Renishaw Ramascope equipped with a
Leica LM optical microscope (a CCD camera). The excitation laser lines with the
wavelengths λex=325 and 532 nm are used in this investigation, and calibration is
conducted using a Si wafer.
3.5 Electrical properties
3.5.1 Electrochemical impedance spectroscopy (EIS) analysis
EIS technique is used to determine the resistive and capacitive characterizations of
materials. The impedance spectrum can be obtained by changing the frequency over
a small applied amplitude sinusoidal alternative current excitation signal [105]. In
this thesis, EIS measurements are carried out to study the electrical conductivities
of the as-synthesized samples and fuel cell devices. The electrical conductivities of
materials and SOFC devices are measured in air or air/H2 atmosphere by an
electrochemical workstation (CHI660B, Chen Hua Corp.) under open circuit
conditions.
3.5.2 Direct-current (DC) polarization analysis
The electronic conductivity is tested using a Hebb-Wagner ion-blocking cell by
means of the DC polarization approach. The DC measurement is performed on a
digital micro ohm meter (KD2531, Kangda Electrical Co., Ltd.) with a voltage
range of 0.01-1.0 V. The Gamry Reference 3000 instrument is employed to analyze
the electrochemical performance of materials and fuel cell pellets. The measured
frequencies range from 0.01 Hz to 1 MHz under a bias voltage of 10 mV. The
Doctoral thesis / Yanyan Liu
27
electrical conductivity (𝜎) for both materials and devices can be calculated using
the equation:
𝜎 =𝐿
𝑅𝐴 (3.1)
Here, 𝜎 is the total electrical conductivity. L presents the thickness of the measured
pellets. R is the total resistance and A is the active area of the pellets. At the initial
stage, the current is raised from the mixed e-/O2- movements under the Hebb-
Wagner cell polarization condition. Subsequently, the oxygen molecules are
consumed at the inner Au/Pr-CeO2 interface with decreasing chemical potential.
Finally, the constant electric current is acquired under the built-in ion-blocking
steady-state condition. In this case, the electronic conductivity can be evaluated
from the obtained current (I) versus time (t) curves.
3.6 Fuel cell performance measurements
The fuel cell performance measurements are conducted for pellet-size devices
connected to a programmable electronic load (ITECH8511, ITECH Electrical Co.,
Ltd.). Hydrogen is supplied to the anode side with flow rates of 120~130 ml min-1
at a pressure of 1 bar, while the cathode side is supplied with air at a similar flow
rate and pressure. The laboratory fuel cell performance testing system is shown in
Figure 3.2. A computerized instrument (IT7000) is employed to measure the I-V
characteristics of a single fuel cell device and corresponding power densities can
be calculated by multiplying voltage and current. The scanning rate for recording
experimental data is set to 0.05~0.1 A s-1.
In this testing system, several potential random factors might impact the accuracy
of the experimental results. One uncertainty comes from the difference between the
set fuel cell operating temperature (determined by a thermocouple) and the actual
cell pellet temperature (non-contact between the pellet and thermocouple). As
discussed in previous sections, the temperature fluctuation can influence the
thermodynamic and electrochemical properties, e.g. ionic conductivity of the solid
electrolyte, thus leading to imprecise fuel cell performance. The experimental error
from the measured furnace temperature by the thermocouple is +/- 5 .
Two different kinds of gas flow meters are employed in this work, which might
also contribute to experimental uncertainty. For the initial experiments, a rotameter-
style (Sho-RateTM 1355) is used to record the H2 flow rates (80~120 mL min-1)
and the air flow rates are not recorded. In the later experiments with an updated gas
flow controller (Bronkhorst F-201CV, for both H2 and air), however, the H2 flow
rates of 120~130 mL min-1 are required to obtain the similar results to the earlier
Doctoral thesis / Yanyan Liu
28
experiments. The desired air flow rate range of 120~130 mL min-1 also is monitored
for the consistent performance outcomes with the previous experimental data. An
accuracy of +/- 2 % for the F-201CV is provided by the instrument supplier. More
details on the gas flow measurements can be found in Xia [158] and Afzal [159].
Some other uncertainty factors might come from the material preparation and fuel
cell fabrication processes, such as the purity of raw materials and weighting
instrument errors. Additionally, some slight heterogeneous particle size and
compositions of composites might influence the sample properties and fuel cell
performance. Repeated experiments (at least three for each set of data) are strictly
required for the material preparation and fuel cell measurements to minimize the
uncertainty and obtain reproducible experiment data.
Figure 3.2 Schematic diagram of fuel cell performance testing setup.
Doctoral thesis / Yanyan Liu
29
Chapter 4
Superionic conductivity of triple-doped ceria for SOFC
This chapter presents a novel approach for modifying the surface properties of SDC
with Pr3+ and Nd3+ dopants, resulting in a triple-doped material. Pr3+ and Nd3+ are
selected due to their potential in extending the electrolytic domain boundary to
lower oxygen partial pressure as well as to enhance the catalytic activity for ORR.
For this purpose, the mechanisms of the enhanced ionic conduction and redox
capability are analyzed. This chapter is based on the published article Paper I.
4.1 Material characterizations
4.1.1 Phase and microstructure properties
Figure 4.1a displays the XRD patterns of mono-cation doping (i.e. SDC) and tri-
cation doping (i.e. PNSDC) ceria oxides. According to JCPDS 34-0394, the
diffraction peaks of the two samples could be assigned to the cubic ceria structure.
A slight shift of diffraction peaks to lower Bragg values can be observed in the
XRD patterns of PNSDC sample compared to SDC. This can be attributed to the
lattice expansion of PNSDC owing to larger ionic radius of Pr3+ ion (i.e. 1.126 Å)
and Nd3+ ion (i.e. 1.109 Å) than Ce4+ (i.e. 0.97 Å) [106]. The lattice constants are
calculated from XRD patterns, which are 5.414 and 5.415 Å, respectively, for SDC
and PNSDC samples. Both of them are higher than that of undoped ceria with a
reported lattice constant of 5.411 Å [104]. The particle sizes of SDC and PNSDC,
according to Scherrer’s equation, are estimated as 55 nm and 61 nm, respectively
[107]. The estimated particle size of PNSDC is approximately 70 nm as shown in
SEM and TEM images (Figure 4.1b). The observed results are consistent with the
calculated data according to XRD results, implying that the as-prepared PNSDC
nanoparticles by the co-precipitation method are homodisperse. The EDS element
mapping analysis (see Figure 4.2) shows that Pr3+ and Nd3+ elements are
successfully distributed onto the SDC scaffold. Therefore, a conclusion can be
Doctoral thesis / Yanyan Liu
30
summarized that Pr3+/Nd3+ doping-ions have been successfully introduced into the
SDC phase without the formation of a second phase.
Figure 4.1 a) X-ray diffraction patterns of SDC (black line) and PNSDC (red
line); b) SEM and TEM micrographs for PNSDC powder.
4.1.2 Ultraviolet-visible (UV-vis) Raman scattering analysis
UV-vis diffuse reflectance spectroscopy technique is employed here to analyze the
surface coordination property and the relationship of absorbance and wavelength
for the obtained samples, shown in Figure 4.3a. A strong absorption can be
observed in the UV section of < 400 nm, typically at the wavelength of 325 nm, for
SDC and PNSDC samples in the shown absorption spectra. In the visible region of
> 400 nm, a weak absorption is observed in the PNSDC line, while SDC absorption
spectrum exhibits a relatively lower absorbance, such as at the wavelength of 532
nm. It could be speculated that the PNSDC sample has an increased oxygen
vacancy concentration compared to SDC due to the replacement of Pr3+/Nd3+ ions
to Ce4+ ions in PNSDC. A related study on the relationship of oxygen vacancy and
UV absorption spectra has been conducted by Sanna et al [50]. As shown in Figure
Doctoral thesis / Yanyan Liu
31
4.3a, the intensity of the absorption for both SDC and PNSDC samples gradually
decreases with the increase of wavelengths. This gives a hint that the different
information from various sampling depths can be collected by means of changing
excitation laser lines using Raman spectroscopy technique. It provides a solution to
analyze the oxygen vacancy information on surface and bulk of samples by
changing the different excitation laser lines (e.g. 325 nm and 532 nm).
Figure 4.2 SEM image of PNSDC and corresponding EDS element mapping.
As reported in the literature, the UV-visible Raman spectroscopy technique is
employed to analyze the surface vibrational mode of materials, while the lattice
structural characteristics are also investigated [108,109]. Figure 4.3b and 4.3c
show the Raman spectra of the SDC and PNSDC samples using excitation laser
wavelengths (λex) of UV (λex=325 nm, Figure 4.3b) and visible (λex=532 nm,
Figure 4.3c). In a fluorite-type CeO2, the symmetrical Ce-O stretching vibration
(denoted as F2g), indexed to surrounding 460 cm-1 with an excitation laser
wavelength of λex=532 nm, plays a dominant role in Raman active modes [109–
112]. In this investigation, the F2g modes for both SDC and PNSDC with a peak
located at vicinal 461 cm-1, marked as peak α in the images, exhibit a shift to lower
energy compared to the reported results for CeO2 [113]. It is worth noting that a
new characteristic peak β, can be observed in the vicinity of 554 cm-1 for both SDC
and PNSDC in Figure 4.3b [113]. This mode indicates the formation of oxygen
vacancies owing to the substitutions of Ce4+ ions by Sm3+, Pr3+ or Nd3+ ions.
Doctoral thesis / Yanyan Liu
32
Compared to the SDC spectrum, the Raman feature lines of PNSDC display a
broader and more asymmetrical characteristic for both F2g mode and the one
induced by oxygen vacancies. It demonstrates a higher dopant concentration and
more oxygen vacancies in PNSDC after the secondary doping Pr3+/Nd3+ ions [114].
Figure 4.3 a) UV-vis diffuse reflectance spectroscopy of SDC and PNSDC; UV-
visible Raman spectra of SDC and PNSDC with an excitation laser wavelength b)
λex=532 nm and c) λex=325 nm; d) Schematic description for Raman scattering by
varying the ultraviolet (UV, λex=325 nm) and visible (λex=532 nm) excitation laser
lines to collect the surface and bulk information.
Figure 4.3d is a diagrammatic drawing showing the relationship of the ultraviolet
and visible excitation laser wavelengths (in this case, λex=325 and 532 nm,
respectively) with the sites of oxygen vacancy defects, e.g. inside bulk or on the
surface of the samples. Using a strong excitation laser with a wavelength of 325
nm, the surface properties of samples are detected from the escaped scattering light
while the major amount of the excitation laser is absorbed. Therefore, the surface
Doctoral thesis / Yanyan Liu
33
oxygen vacancy can be determined via UV Raman lines. Accordingly, the bulk
properties of PNSDC and SDC samples are obtained using a relatively weak
excitation laser, e.g. a wavelength of 532 nm [108,109,115]. Figure 4.3c exhibits
a significantly intensive peak for PNSDC (peak β) in the vicinity of 554 cm-1 using
a UV excitation laser of 325 nm, compared to that of SDC sample.
Further evidence is gathered to support the replacement of Ce4+ ions using Pr3+/Nd3+
ions, particularly on the surface regions of the PNSDC sample. As stated previously,
peaks β in Raman spectra reflects the oxygen vacancy information and peak α
represents the F2g mode information of the doped ceria with a fluorite-type structure.
The intensity ratio between the two characteristic peaks (written as Aβ/Aα) is
calculated by integration of individual peak area. The results of Aβ/Aα reflect the
oxygen vacancy concentration and indeed the defect sites combining with measured
depths of PNSDC and SDC [116]. By computation, the values of Aβ/Aα for PNSDC
and SDC are approximately 0.298 and 0.301, respectively, for the case using
excitation laser wavelength of 532 nm (Figure 4.3b). The similar results
demonstrate no apparently structural change occurring inside the bulk of PNSDC
sample after secondary Pr3+/Nd3+ doping. However, an obvious change is observed
when using the UV Raman excitation laser, i.e. 325 nm, as shown in Figure 4.3c.
The calculated Aβ/Aα value is about 1.512 for PNSDC, while 1.294 is obtained from
the calculation for the peak areas of the SDC sample. The marked difference
implies a higher concentration of the oxygen vacancy in PNSDC compared to that
of SDC. Thus, one can conclude that the Pr3+ and Nd3+ ions are doped on the surface
sites of SDC particles.
4.1.3 XPS analysis
As shown in Figure 4.4a, the survey spectra exhibit the characteristics of various
elements, i.e. Sm, Ce, Pr, Nd, C, and O in the PNSDC. The individual XPS peaks,
representing the O 1s, Pr 3d, and Ce 3d, are tested to identify the chemical states
on the surface of PNSDC. The standard Gaussian functions are employed to
analyze these relevant multiple-component XPS featuring peaks.
Figure 4.4b displays the O 1s core-level spectra related to oxygen species,
including surface chemisorbed oxygen or oxygen defects (i.e. O−, Oα) with a
binding energy of 531.0-532.6 eV and lattice oxygen with binding energy (Oβ) at
the range of 528.8-529.4 eV [117]. According to literature, the bonds associated
with Ce-O-doping cation are generated when the doping ions are introduced into
the host ceria lattice, thus resulting in more mobile oxygen species to facilitate the
Doctoral thesis / Yanyan Liu
34
Figure 4.4 a) XPS survey spectra of SDC and PNSDC samples; b) XPS O 1s
Spectra of SDC and PNSDC samples; c) XPS Ce 3d Spectra of SDC and PNSDC
samples; d) XPS Pr 3d Spectra of PNSDC sample.
migration of lattice oxygen species from the internal sites to the surface regions
[118]. The Ce 3d spectra of SDC and PNSDC are shown in Figure 4.4c.
Both spectra for the two samples give information that the Ce3+ and Ce4+ ions are
coexistence on the surface. Generally, the Ce 3d core-level spectrum can be
deconvolved into several different spin-split doublets aiming at analyzing the
fraction for the oxidation states of Ce3+/Ce4+ ions and thus in-depth understanding
the effects of doping ions, i.e. Sm3+/Pr3+/Nd3+, on the surface properties of PNSDC
[119,120]. According to the spectra analysis on Figure 4.4c, the percentage
composition of Ce3+ and Ce4+ species is calculated roughly using the deconvolution
peaks relevant to (v0, v′, u0, u′) and (v, v′′, v′′′, u, u′′, u′′′), respectively. The
estimated results exhibit an increasing Ce3+ ion concentration for 2% higher than
that of SDC. It is well known that the low-valance Ce3+ ions are closely related to
the oxygen vacancy, and thus a higher oxygen vacancy concentration is expected
for the PNSDC sample, especially on the surface of the sample [121,122]. As
Doctoral thesis / Yanyan Liu
35
reported previously [123,124], the Pr 3d spectra could be deconvoluted into the
distinguishable peaks, mirroring the information associated with the Pr3+/Pr4+
oxidation states. From Figure 4.4d, two featuring peaks for Pr4+ ions, i.e. 3d5/2 and
3d3/2, are detected respectively at surrounding 931.6 and 950.2 eV, while the
characteristic peak pairs for Pr3+ ions are observed at around 933.3/928.4 eV and
953.1/947.2 eV for 3d5/2 and 3d3/2, respectively. As observed, the PNSDC sample
has coexistence of Pr3+/Pr4+ oxidation states with much larger distinguishing peaks
of Pr3+ ions than that of Pr4+ ions, confirming a dominant role of Pr3+.
4.2 Electrical properties
EIS is carried out here to analyze the electrical property of the SDC and PNSDC
samples. In general, the intrinsic charge carriers play a dominant role in the
conduction of ionic conductor measured in air atmosphere. In a typical ion-
conducting case, the grain, grain boundary and electrode polarizations make main
contributions to the EIS including one arc at high frequency region, one arc at
middle frequency region and a following tail at low frequencies, respectively. In
some practical measurements, the high-frequency arcs might not show up due to
the not enough high frequency limited by the testing setup or operating temperature
[125].
Figure 4.5 shows the comparative Nyquist plots for SDC and PNSDC. All the plots
for both samples display a middle-frequency arc referring to grain boundary
resistance and low-frequency tail representing the electrode polarization processes.
Both high-frequency arcs fail to appear due to the limited frequency range of the
testing equipment. It is obvious that all the sizes of EIS decrease as the applied
temperature increases. ZVIEW software is employed to fit these EIS results. As
fitting data, the grain capacitors have a value level of around 10-12 F. The capacitors
for grain boundaries range from 10-10 F to 10-8 F, while the capacitors for electrode
polarization exhibit values approximately 10-4 F for both SDC and PNSDC samples.
These results are consistent with the values reported previously [126,127]. In
contrast, the PNSDC sample has a sharply decreasing grain-boundary arc at the
various temperature as compared to SDC. Figure 4.6a presents the temperature-
dependent conductivities of Arrhenius plots for PNSDC calculated by the EIS
measurements and the comparative results for SDC are also provided. As estimated,
the PNSDC sample exhibits an enhanced conductivity as 0.125 S cm-1 at 600 ,
while the conductivity of SDC is 0.0114 S cm-1 at the same temperature. The
activation energy is determined according to the Arrhenius plots: 0.41 eV for
PNSDC, lower than that of SDC, 0.65 eV. These results are comparable with the
reported data in the literature [128]. The different conduction mechanisms for
Doctoral thesis / Yanyan Liu
36
Figure 4.5 Electrochemical impedance spectra of SDC and PNSDC measured in
air at a) 400 ; b) 450 ; c) 500 ; d) 550 ; e) 600 and f) the empirical
equivalent circuit to analysis the ionic conductivity of SDC and PNSDC materials
in air.
PNSDC material could be speculated that the reasons are the significant
improvements in the aspects of activation energy and conductivity compared to
SDC.
To identify the possible electronic conduction of these doped ceria samples, the
Hebb-Wagner ion-blocking technique is employed here with an applied potential
range of 0.01-1.0 V. During the measuring process, the mobility of oxide ions and
electrons contribute to the current behaviors at the initial polarization stage. Due to
Doctoral thesis / Yanyan Liu
37
the ion transfer blockage, the chemical potential exhibits a rapidly decreasing trend
at the interfaces of Au and PNSDC sample as the testing time prolongs. The
electronic conduction property for PNSDC sample is presented in Figure 4.6b in
terms of current versus time (I-t) curves at the cooling temperature of 600, 550, 500
to 450 . At 600 , the electronic conductivity is approximately determined as
6.4×10-4 S cm-1, which is negligible compared to the value of total conductivity as
0.125 S cm-1. In conclusion, the enhanced conduction of PNSDC is attributed
primarily to ion transport, especially via conductivity of grain boundaries, and the
Pr3+/Nd3+ doping is favorable for surface ion incorporation to result in an improved
conductivity as well lower activation energy.
Figure 4.6 a) Temperature dependence of total conductivity and Arrhenius plots
for SDC and PNSDC. The inset represents area specific resistance (ASR) versus
temperature of SDC and PNSDC; b) current (I) versus time (t) curves obtained at
a temperature range of 450 and 600 at an interval of 50 to determine the
electronic conductivity of PNSDC sample.
4.3 Fuel cell performance
The fuel cell performance for cell I (SDC electrolyte) and cell II (PNSDC
electrolyte) are displayed in Figure 4.7a. The OCV for SDC cell is 0.922 V while
that of PNSDC device reaches up to 1.021 V at 550 , indicating minimal
electronic leakage for the PNSDC electrolyte. Correspondingly, the maximum
power density of cell II is 710 mW cm-2, which is much higher than the power
density of 490 mW cm-2 for SDC device. The evident enhancement can be
attributed to the reduced resistances of PNSDC electrolyte and PNSDC/NCAL
composite electrode resistances. Good durability is recorded for cell II under
operating conditions of 80 mA cm-2 discharge current density at 530 during 13
h, as shown in Figure 4.7b. Figure 4.8 presents the EIS results for analyzing the
Doctoral thesis / Yanyan Liu
38
Figure 4.7 a) Typical I-V and I-P curves of fuel cells based on SDC and PNSDC
electrolyte, respectively, the mixed NCAL and electrolyte materials were used as
the electrode material. The measurements were performed at 550 with H2 and
air as the fuel and the oxidant, respectively; b) Durability test based on Cell II:
Ni-foam/PNSDC+NCAL (anode)|PNSDC (electrolyte)| PNSDC+NCAL
(cathode)/Ni- foam at a constant discharge current density 80 mA cm-2 operated
at 530 .
Figure 4.8 Electrochemical impedance spectra of the symmetric cell using SDC
and PNSDC samples as electrolyte measured under a) air and b) H2/air
conditions at 550 .
fuel cell kinetic behaviors for both cells I and II measured at 550 . An empirical
equivalent mode, denoted as Ro(R1-CPE1) (R2-CPE2), is used here to represent the
O2- mobility during the fuel cell process [129]. As shown for both cases of the
applied air and fuel cell conditions, the ohmic resistance as determined from the
high-frequency intercept (Ro) is markedly reduced for the PNSDC device as
compared to the SDC device, which can be linked to enhanced oxygen ion
conduction for PNSDC [130]. Simultaneously, the oxygen ion migration from the
Doctoral thesis / Yanyan Liu
39
Figure 4.9 Schematic diagrams representing a) the overall O2‐ transfer with the
assistance of reduction conversation of Pr3+/Pr4+ ions in the fuel cell; b) the
cathode reaction process and c) brief routes for the O2‐ surface and bulk transfer
and conduction process in PNSDC electrolyte.
cathode through the electrolyte is improved, thus inducing a reduced electrode
polarization loss. The improved surface O2- conductivity is associated with the
reducing grain boundary of PNSDC. Figure 4.9 shows the schematic illustration
exhibiting the transfer routes of oxygen species in the PNSDC electrolyte. As
shown, the oxygen species are transferred via a redox process relying on the ion
chains [131,132]. During the redox reaction process, the O2 molecules from applied
air are initially activated and then reacted with the electrons from the connecting
external circuit. Furthermore, the electrons exchange between redox couple of
Doctoral thesis / Yanyan Liu
40
Pr3+/Pr4+ ions can facilitate the O2- transport in PNSDC [131,132]. This process is
described using the following formula:
Pri3++Prj
4+↔Pri4++Prj
3+ (4.1)
At the anode, the Pr4+ ions are reduced to Pr3+ by H2. Meanwhile, the electrons
escape from the reaction sites with a specific rate and their mobility is limited
largely by the intrinsic defects of PNSDC. At the cathode, the trivalent Pr ions
would be re-oxidized to tetravalent state and an entire redox couple, i.e. Pr3+/Pr4+,
is realized, where oxygen ions are involved selectively [131,132]. In this case, a
high OCV of 1.021 V at 550 is obtained arising from the characteristic redox
couple Pr3+/Pr4+ against the possible reduction of Ce4+ in this triple-doped ceria
material. To further clarify the mechanism related to the enhanced redox process
through the variable valence of single-Pr ion, a surface Pr-doping solely ceria is
developed to serve as the electrolyte of SOFCs, performing fuel cell measurement
for comparison (in Chapter 5).
As reported in the literature [133, 134], the dominant polarization loss in the low
operating SOFC system is the ORR process, and the improved cathode ORR
process for PNSDC is favorable for improved power output in the SOFC device.
The generated O2- ions at the cathode are easily transferred via both bulk and
surface routes of PNSDC in cell II. Thus, the efficiency of ion migration is
enhanced and fuel cell performance is improved. As reported in a recent study [129],
the lithiated transition metal oxide confirmed good catalytic properties for HOR as
anode and ORR as the cathode. In this study, the NCAL is employed as both the
anode and cathode to reinforce the HOR and ORR properties and the combining
PNSDC or SDC phases can extend the triple phase boundary interfaces, resulting
in the enhanced electrochemical performances [135–138]. Furthermore, the
polarization losses from both electrolyte and electrode can be alleviated owing to
the excellent electrochemical catalysis of Pr ions. Therefore, the SOFC device
using PNSDC demonstrates superior electrochemical performance.
4.4 Chapter summary
An effective approach is developed to significantly improve the oxygen ionic
conductivity by more than one order of magnitude while minimizing the redox
instability for the doped ceria-based materials. A two-step co-precipitation strategy
is employed here to realize the varied doping sites of ceria: Sm3+ in bulk and
Pr3+/Nd3+ on the surface. The grain-boundary resistance of PNSDC sample has been
verified largely decreasing with a conductivity of 0.125 S cm-1 at 600 .
Additionally, the Pr3+/Nd3+ ions are found to facilitate the surface oxygen species
Doctoral thesis / Yanyan Liu
41
exchange and to reinforce the electrochemical kinetic processes, thus improving
the entire fuel cell performance. In this investigation, the PNSDC-based SOFC
device exhibits excellent performance, e.g. a peak power density of 710 mW cm-2
at the operating temperature of 550 .
Doctoral thesis / Yanyan Liu
42
Doctoral thesis / Yanyan Liu
43
Chapter 5
Multi-component rare-earth doped ceria for EFFC
In this chapter, single-, double- and triple-doped approaches are proposed for the
modification of doped ceria materials aiming at applications for EFFC technology.
Single-phase Pr-doped ceria with mixed ionic and electronic conduction is
investigated as a core component of EFFC devices. Double- and triple-doped ceria
materials combined with semiconductor served as homogenous semiconductor-
ionic composites for EFFCs. Fundamental material properties and fuel cell
performance related these materials are studied systematically. Both experimental
results and data analysis are included, as reported in Paper II, III, V and VI.
5.1 Single-phase Pr-doped ceria for EFFC
5.1.1 Structural and morphology characteristics
The XRD patterns of the Pr-CeO2 crystalline powder synthesized via the
hydrothermal method in comparison with Pr-free CeO2 are shown in Figure 5.1.
Results for the two samples are related to a cubic fluorite structure with a space
group of 𝐹𝑚3𝑚 (Ref. standard CeO2 JCPDS card: NO.04-0593). A slight left shift
is observed in the Pr-CeO2. In addition, the sharp diffraction peaks indicate the
presence of nanoparticles with regular crystallinity. No other diffraction peaks of
impurities are detected within the detectability of XRD, demonstrating that a single-
phase Pr-CeO2 sample with a nominal composition of Pr0.1Ce0.9O2-δ is synthesized.
SEM and TEM images are performed to reveal overall particle morphology and
qualitatively investigate the size-distribution for the synthesized CeO2 and Pr-CeO2
particles. Figure 5.2 illustrates the representative SEM and TEM images of as-
synthesized CeO2 (a and b) and Pr-CeO2 (c and d) samples. As shown, Pr-free CeO2
Doctoral thesis / Yanyan Liu
44
powder is comprised of aggregated particles to form rod crystallites. However, the
Pr-doped CeO2 exhibits homogeneous particles. The small nano-crystal of the Pr-
CeO2 sample are expected to improve fuel cell performance in comparison to that
of CeO2 owing to a higher specific surface area.
Figure 5.1 X-ray diffraction patterns of as-synthesized Pr-free CeO2 and Pr-CeO2
samples.
5.1.2 XPS analysis
XPS analysis is carried out to distinguish the chemical bonding states of elements
on the surface of as-prepared CeO2 and Pr-CeO2 samples. The obtained spectra and
corresponding fitting results are presented in Figure 5.3. The survey spectra in
Figure 5.3a clearly reveal the characteristics of C, O, Ce and Pr elements. To
further identify the chemical states of the relevant elements, the O 1s, Ce 3d and Pr
3d core-level spectra are addressed using XPSPEAK 4.1 software. Figure 5.3b
illustrates the Pr 3d spectra. The characteristic peak pairs of Pr3+ 3d5/2 (located at
around 928/933 eV) and 3d3/2 (located at around 947/956 eV) are observed in the
fitting spectra. The peaks of ~931 and 952 eV can be assigned to Pr3+ 3d5/2 and 3d3/2,
Doctoral thesis / Yanyan Liu
45
Figure 5.2 Representative SEM micrographs for as-synthesized a) CeO2 and c)
Pr-CeO2 samples; TEM images of b) CeO2 and d) Pr-CeO2 samples suffering
hydrothermal process.
respectively. The fitting results demonstrate that both the oxidation states of Pr3+
and Pr4+ are detected in the Pr-CeO2.
As reported in the literature [139,140], the reduction of Pr4+ to Pr3+ can create
oxygen vacancies. In this case, the mixed oxidation states of Pr3+ and Pr4+ can
promote the ionic conductivity and facilitate to enhance the electrochemical
performance for Pr-CeO2. Figure 5.3 c and d show the O1s core-level spectra of
CeO2 and Pr-CeO2, respectively, with the two feature peaks relevant to the lattice
oxygen (Oα) with the binding energy range of 528-530 eV and chemically absorbed
surface oxygen species or oxygen defects (Oβ) located at around 530-533 eV[117].
Generally, the Ce-O-Pr bonds can be formed once Pr ions are incorporated into the
CeO2 lattices. Subsequently, the electro-negativity has slight differences for Pr and
Ce ions, and more changeable oxygen leads to the facilitated migrations of oxygen
from internal sites to the surface[118]. As calculated, the Oβ/Oα value of Pr-CeO2
Doctoral thesis / Yanyan Liu
46
Figure 5.3 XPS spectroscopy: a) survey spectra of CeO2 and Pr-CeO2 samples;
b) Pr 3d spectra for Pr-CeO2; c) O 1s spectra of CeO2; d) O 1s spectra of Pr-
CeO2; e) Ce 3d spectra of CeO2; and f) Ce 3d spectra of Pr-CeO2.
(~1.24) is higher than that of CeO2 (~1.21), indicating an increase of surface oxygen
vacancies in Pr-CeO2 sample. The Ce3+ and Ce4+ ions are detected in coexistence
on the surface of both CeO2 and Pr-CeO2, as presented in Figure 5.3 e and f. The
spectra deconvolution with several peaks in CeO2 and Pr-CeO2 are assigned to the
Doctoral thesis / Yanyan Liu
47
Ce 3d5/2 (vo, v and v’) and 3d3/2 (uo, u, u’ and u’’) [118,141]. In this case, the peak
groups of (uo, vo) and (u, u’ u’, u’’, v, v’) are attributed to Ce3+ and Ce4+ species,
respectively [108,118]. It can be deduced from the resultant spectra that Ce3+ and
Ce4+ ions coexist on the surface of CeO2 and Pr-CeO2.
5.1.3 Raman scattering analysis
Figure 5.4 Raman Spectroscopy of CeO2 and Pr-doped CeO2 samples using an
excitation wavelength of 325 nm.
The Raman reflectance spectra with an excitation laser wavelength of 325 nm for
CeO2 and Pr-CeO2 samples are presented in Figure 5.4. The dominant Raman
active mode is observed at around 460 cm-1 for the CeO2 sample, while both this
mode and a second one at 554 cm-1 are seen for the Pr-CeO2 sample. The peak at
around 460 cm-1 is attributed to the symmetrical oxygen lattice Ce-O stretching
vibrational mode (F2g) in a cubic fluorite-type CeO2 lattice [140,142,143], which
exhibits a decreasing energy intensity for Pr-CeO2 compared to that of CeO2.
Several factors, e.g. strain, defects, and phonon confinement, can induce the
variation of F2g peak. In this case, the slight shift in the F2g mode of Pr-CeO2 results
from the larger ionic radius of Pr3+ ions (1.013 Å) as compared to host Ce4+ ions
Doctoral thesis / Yanyan Liu
48
(0.92 Å) [144]. Additionally, the oxygen vacancies also contribute to the shift of
F2g peak [145]. The larger ionic radius of Pr3+ (as compared to Ce4+) contributes to
the lattice expansion [146]. The emerging feature peak, at around 554 cm-1 for the
Pr-CeO2 sample is indexed to the formation of oxygen vacancies induced by the
substitution of Pr3+ in CeO2 surface lattice, while such defect-related peaks are not
observed in Pr-free CeO2 sample. This finding indicates that the defects are
generated due to the introduction of Pr ions into the ceria crystal structure,
compared with the CeO2 spectrum without the defect mode [146]. The oxygen
vacancies are formed in the CeO2 lattice, similar to Gd-doped CeO2, by doping with
Pr3+ cations:
Pr2O32CeO2→ 2PrCe
′ + VO·· + 3OO
x (5.1)
As shown in Equation 5.1, two Pr3+ cations substitute two Ce4+ ions to generate an
oxygen vacancy to balance the charge based on the charge compensation
mechanism [147,148]. It has been reported that the asymmetry on the high-
frequency side of the Raman line in the spectrum for Pr-CeO2 is contributed from
the existence of a probable surface mode at 480 cm-1 [148]. This is ascribed to the
large defective structure and disorder in the ceria lattice.
Figure 5.5 a) Nyquist plots for Pr-CeO2 sample measured at 600 under air
atmosphere; b) Arrhenius plot and corresponding temperature dependence of
total conductivity for Pr-CeO2 sample.
5.1.4 Electrical conductivity
The electrical behaviors of the as-prepared Pr-CeO2 sample are examined by EIS
in air atmosphere operated at a temperature range between 500 and 600 with
Doctoral thesis / Yanyan Liu
49
an interval of 25 . Figure 5.5a demonstrates the typical Nyquist plot at 600 . It
is well known that the electrical properties are well correlated to the operating gas
atmosphere. Generally, the intrinsic charge carrier is dominant to the conductivity
in the air environment [125]. In Figure 5.5a, the Nyquist plots of Pr-CeO2 sample
demonstrate a typical medium frequency arc and low-frequency tail. In this case,
the arcs in the high-frequency range could not be detected. Prior to further EIS
analysis, a possible electronic conductivity needs to be considered since the
existence of a Pr impurity band within band gap of ceria facilitates electron hopping
in the Pr-CeO2 structure [149,150]. The estimated electronic conductivity of Pr-
CeO2 using Hebb-Wagner ion-blocking technique is approximately 0.022 S·cm-1 at
600 .
To understand the electrochemical kinetic behavior in detail, EIS simulations for
impedance plots are conducted using Zview software package. The empirical
equivalent circuit model of R0(R1-QPE1)(R2-QPE2), as shown in the inset of Figure
5.5a, is employed. In this equivalent model, R0 is assigned to ionic conduction
indexing to the intercept of the EIS curve with the Z’-axis at high-frequency range
(>104 Hz). R1 is related to the oxygen transfer at the Ag/Pr-CeO2 interface under
the frequency range of 102-104 Hz. The oxidation/reduction of oxygen processes
are associated with the resistance of R2 in the low-frequency range of 10-2-102 Hz.
QPEs are the constant phase element to express the ‘depressed’ arc in consideration
of any inhomogeneity in the measured material [150–152]. The total conductivity
values for Pr-CeO2 are estimated as 0.36, 0.16 and 0.037 S cm-1 at 600, 550 and
500 , respectively. Compared with the above-calculated results, the electronic
conductivity (~0.022 S·cm-1) is one order of magnitude lower than the ionic
conductivity (~0.338 S·cm-1) for Pr-CeO2 at 600 . The temperature-dependent
conductivities for Pr-CeO2 sample are presented in Figure 5.5b in the form of an
Arrhenius plot. The activation energy of Pr-CeO2 is estimated to be 1.3 eV,
according to the Arrhenius relationship:
σ = (σ0/T)exp(−Ea/kBT) (5.2)
where σ and σo refer to the conductivity and pre-exponential constant, respectively,
Ea is the activation energy, kB is the Boltzmann constant and T is the operating
temperature.
The unexpected high conductivity of Pr-CeO2 could be contributed to two
dominating factors: i) rapid oxygen ion transfer and surface exchange resulting
from enhanced surface oxygen vacancy concentration; generally, the oxygen
species can be quickly absorbed and transferred through oxygen vacancies, so
Doctoral thesis / Yanyan Liu
50
higher ionic conductivity can be expected with more vacancies; ii) promoted
electronic conduction arising from impurity Pr band, acting as an electron
conduction channel, within band gap of ceria and faster oxygen exchange [152,153].
In parallel, compared to the purely ionic conductors, the mobile electronic carriers
benefit to the catalytic activity on the surface of materials [152,154]. It can be
speculated that the electron conduction and ion transport have a strong correlation
to tremendously enhance the coordinated conductivity instead of the simple sum of
both contributions. It has verified that the ionic conductivity was significantly
enhanced by compositing with semiconductors, insulating conductors, or with
carbonates [36,155]. Analogously, the single-phase materials with electronic and
ionic conduction have similar synergistically enhanced conductivity to those two-
or three-phase composites.
5.1.5 Fuel cell performance analysis
Figure 5.6 I-V-P characteristics of fuel cell devices based on a) the CeO2 and Pr-
CeO2 membrane separately measured at the temperature of 550 ; b) Pr-CeO2
operated at various temperatures of 500, 525, 550 and 600 , respectively.
The current-voltage (I-V) and current-power (I-P) characteristics for symmetrical
fuel cells using CeO2 (as a comparison) and Pr-CeO2 as the membrane layer,
respectively, are presented in Figure 5.6a. It can be observed that both OCVs for
the fuel cells reach up to 1.0 V at 550 . The high OCV values indicate that the
fabricated devices realized the fuel cell function without electronic current leakage.
At this operating temperature, an appreciable peak power density of 744 mW cm-2
is achieved for the EFFC device using Pr-CeO2, which is superior to that of CeO2
of 497 mW cm-2. Such enhancement of the power output reflects a higher
conductivity of the Pr-CeO2 electrolyte, compared to the CeO2 electrolyte.
Doctoral thesis / Yanyan Liu
51
Figure 5.7 Schematic diagrams representing a brief route for the O2- mobility in
fuel cell configuration and detailed O2- transfer with the assistance of redox
Pr3+/Pr 4+ couple on the surface and bulk of Pr-CeO2 electrolyte in the vicinity of
oxygen applied side.
To further investigate the property for the Pr-CeO2 sample, the fuel cell
measurements are performed at various operational temperature, as shown in
Figure 5.6b. It demonstrated that the maximum power densities of 776 and 471
mW cm-2 are obtained at 600 and 525 , respectively. Significantly, even the
operating temperature down to 500 , a considerable peak power density of 296
mW cm-2 is still obtained.
The relatively high output power density for Pr-CeO2 is based on the following
factors: i) enhanced ionic conductivity of Pr-CeO2 membrane via Pr surface dopant;
ii) promoted ion conduction strongly correlated to electronic conductivity due to
electron hopping between neighboring Pr ions via impurity band conduction under
low oxygen partial pressure; iii) improved electron transfer under high oxygen
partial pressure owing to the redox Pr3+/Pr4+ pair. As reported by Zhu et al. [34], a
thin enriched nickel metal layer (≈ 4-6 μm) formed in the LiNi0.85Co0.15O2-δ layer
on the hydrogen-supplied side with the reduction of LiNi0.85Co0.15O2-δ. On the H2-
contacting side, a Schottky junction was verified to be established by supplying
external voltage and exhibiting a typical rectification effect in such a device. In this
study, the high OCV and good fuel cell output power density indicate that a similar
phenomenon occurs and Schottky junction is established in the H2-contacting side
Doctoral thesis / Yanyan Liu
52
to prevent short-circuiting. The schematic diagrams representing a brief route for
the O2- mobility in fuel cell configuration and detailed O2- transfer with the
assistance of redox Pr3+/Pr4+ couple on the surface and bulk of Pr-CeO2 in the
vicinity of oxygen applied side are presented in Figure 5.7. As shown, at the high
oxygen partial pressure region (O2-contacting side), Pr largely remains as
tetravalent Pr4+ and can be reduced by electrons from the external route during fuel
cell process. O2- ions can be transported relying on redox Pr3+/Pr4+ couple and
oxygen vacancies via bulk and surface routes. With reducing of oxygen partial
pressure, oxygen is released from the lattice with converting Pr4+ to Pr3+ and oxygen
species can be transferred flexibly through the dominative Pr-doped CeO2
membrane. As mentioned above, the fuel cell using Pr-CeO2 membrane is
significantly improved in comparison to that of Pr-free CeO2.
5.2 Rare-earth triple-doped ceria for EFFC
Figure 5.8 SEM micrographs for a) raw material and b) as-prepared LCPN; c)
TEM microstructure and corresponding selected area electron diffraction pattern
(SAED, inset image) for LCPN sample.
Microstructures of the raw material and as-obtained LCPN sample are presented in
Figure 5.8 by means of SEM and TEM images. As shown in the SEM image for
the raw rare-earth material, regular particles are shown to be agglomerated into a
range of micrometer sizes. After calcination, the LCPN sample displays the
relatively homogeneous particles with an estimated size of 40 nm. The
polycrystalline structure of the as-obtained LCPN powder is confirmed by the
SAED pattern, see inset in Figure 5.8c. To further underline the elemental
compositions, EDS experiments are conducted on the raw material and treated
LCPN sample, see Figure 5.9. During the heat-treatment process, the raw rare-
earth carbonates are reacted into rare-earth oxides with a decreasing carbon
Doctoral thesis / Yanyan Liu
53
composition and increasing oxygen species. The rare-earth elements (i.e. Ce, La,
Pr, Nd) and oxygen are confirmed as the dominating compositions of the obtained
LCPN material from the EDS analysis.
Figure 5.9 Typical SEM micrographs of a) raw material and c) LCPN; b) and d)
identified the EDS data with the representative counts of the vertical axis
obtained at the yellow rectangular areas.
Figure 5.10 XRD patterns of LCPN and NCAL samples.
Doctoral thesis / Yanyan Liu
54
Figure 5.10 exhibits the XRD results for the LCPN and NCAL. LCPN pattern can
be assigned to the cubic-fluorite ceria with a slight left-shift of the diffraction peaks
owing to the elements of La3+/Pr3+/Nd3+ with larger ionic radii compared to the Ce4+
ions. No secondary phase is detected by the XRD measurement. The commercial
NCAL is analyzed as a layer structure, indexing to the diffraction peaks of
LiNi0.8Co0.2-oxide (JCPDS, No. 87-1562) [32]. The aluminum element is not
detected due to the limitation of testing [33].
Figure 5.11a shows the electrochemical properties of the device using
LCPN/NCAL composite with a weight ratio of 7:3 measured at a temperature range
of 550-600 at intervals of 25 . All the three EIS patterns demonstrate two
components: a depressed small semicircle at the high-frequency region and another
half-arc at low-frequency range. The equivalent model (see the inset image of
Figure 5.11a) is expressed as Ro/R1(CPE1)/R2(CPE2) and is employed to analyze
the EIS results. The intercept at high frequencies represents the interfacial
resistance of charge transfer, denoted as R1, while R2 can be ascribed to the
resistance of mass transfer, indexing to the low-frequency semicircle [27]. In
general, the reaction kinetics associated with charge and mass transport can be
improved by increasing the operating temperature, which is consistent with the
measured results of Figure 5.11a. Furthermore, the EFFC devices using the
LCPN/NCAL composites with various weight ratios, i.e. 7:3, 6:4, and 3:7 for LCPN
and NCAL are investigated to select the optimum proportion of ion-conducting
phase and electron-conducting phase, as shown in Figure 5.11b. All the devices
are tested at 550 fed by air and hydrogen, exhibiting a similar characteristic EIS
curve. In contrast, the polarization resistances are reducing with the increasing
contents of the electronic phase, i.e. NCAL, reflecting a reduced semicircle in EIS
plots. The increase in electron conductor NCAL might cause possible electrical
leakage. For instance, some leakage might happen for the device using the 6LCPN-
4NCAL composite, exhibiting a slight lower OCV value compared to the pure
LCPN and 7LCPN-3NCAL devices, as shown in Figure 5.12a. According to the
fitting results of EIS, the resistance of charge transfer for the device using 7LCPN-
3NCAL is approximately 0.46 Ω cm2, while it is only 0.23 Ω cm2 for 3LCPN-
7NCAL. However, the resistances for three devices using LCPN/NCAL
composites are reduced due to the additional NCAL, indicating favorable kinetic
reactions for these devices.
Doctoral thesis / Yanyan Liu
55
Figure 5.11 EIS patterns of a) the 7LCPN-3NCAL composite at different
temperatures and b) the different compositions of LCPN and NCAL (7LCPN-
3NCAL, 6LCPN-4NCAL, and 3LCPN-7NCAL) under air/H2 condition at 550 .
All the measurements were performed under air/H2 condition under open circuit
conditions.
Fuel cell performance is investigated based on these fuel cell devices using various
LCPN/NCAL composites and pure LCPN, see Figure 5.12a. The device using a
pure LCPN as the electrolyte displays a high power density of 1,046 mW cm-2
measured at 550 . It indicates that the LCPN material can serve as a good
electrolyte candidate with high ionic conductivity. After composition with a
semiconductor, e.g. 7LCPN-3NCAL composite, the fuel cell device exhibits the
maximum power output (Pmax), reaching 1,187 mW cm-2 at 550 and the OCV
reaches as high as 1.07 V among others. If the electron-conducting phase is
increased to 40 wt.% in the LCPN/NCAL composite, a relatively low power density
is obtained. Further deterioration in performance is exhibited as the NCAL fraction
increases, resulting in a poor OCV (0.4 V) for the 70 wt.% NCAL sample. These
findings indicate that good EFFC performance is dependent upon proper balancing
between the ion-conducting phase, i.e. LCPN, and electron-conducting phase, i.e.
NCAL. It has been verified that a Schottky junction could be established on the
hydrogen supply side, which can block electron flow through the internal layer of
the device [1,15]. Figure 5.12b presents the various power outputs and OCV values
for the devices using different weight ratios of LCPN/NCAL composites. As shown,
a decreasing trend for the OCV of these devices is found with the increase of NCAL
composition. For example, a high OCV of 1.18 V is obtained when the pure LCPN
is used as the electrolyte of SOFC. The device using the LCPN with an addition of
30 wt.% NCAL exhibits an OCV of 1.07 V, while the one using 5LCPN-5NCAL
Doctoral thesis / Yanyan Liu
56
composite displays the OCV for approximately 0.92 V. Indeed, the 7LCPN-
3NCAL composition has a favorable OCV value, and it also shows the best
performance in high sustained power density across a wide range of loads. The
enlarged triple phase boundaries of LCPN/NCAL composite with concomitant
lower polarization loss contributes to the enhanced fuel cell performance.
Figure 5.12 a) The I-V (current density and voltage) and I-P (current density and
peak power density) characteristics of various compositions of LCPN-NCAL
feeding with hydrogen and air at 550 ; b) OCV and maximum power density
values (Pmax) for fuel cells based on LCPN-NCAL in various compositions.
Figure 5.13 Schematic illustration (left) and SEM micrograph (right) of EFFC
device based on the 7LCPN-3NCAL composite after the fuel cell performance
measurements.
Doctoral thesis / Yanyan Liu
57
Figure 5.13 displays the cross-section image for the tested EFFC device using
7LCPN-3NCAL composite and corresponding schematic sketch. As observed,
some structural changes are found in the hydrogen-contacting regions, which could
be attributed to the reduction of NCAL to yield nickel or cobalt metals. Based on
these formed metals, a Schottky junction is built inside the LCPN/NCAL composite,
via contacting with p-type NCAL semiconductor, as verified in literature [15,17,18].
From the SEM image of the EFFC device, both NCAL layers retain a dense
structure even after fuel cell testing.
5.3 Double-doped ceria for EFFC
Figure 5.14 a) I-V and I-P characteristics and b) EIS of EFFC devices using the
SCDC/LSCF composites in various weight ratios; c) I-V and I-P performance for
the conventional SOFC device and the device using pure SCDC membrane.
The typical I-V and I-P curves for EFFC devices with the configuration Ni-
NCAL/LSCF-SCDC/Ni-NCAL based on the various weight ratios of LSCF/SCDC
are presented in Figure 5.14a. All of these fuel cell devices are tested at 550 .
As observed, the best performance is obtained for a weight ratio of 1:1 (LSCF and
Doctoral thesis / Yanyan Liu
58
SCDC) with a maximum power density of 814 mW cm-2. Figure 5.14b shows the
EIS curves of the three fuel cells under H2/air condition in open circuit mode. As
shown, the entire polarization losses as LSCF content increased. The comparative
fuel cell performance for the conventional SOFC device and the device using pure
SCDC membrane are shown in Figure 5.14c. As shown, the conventional device
with the symmetrical configuration of NCAL/SCDC/NCAL, showed a 20% higher
power density (e.g. 368 mW cm-2) as compared to the conventional FC device using
SCDC as electrolyte and LSCF and NCAL as cathode and anode respectively (e.g.
300 mW cm-2).
5.4 Fuel cell performance comparison
Table 5.1 Fuel cell performance comparison based on three types of doped ceria
materials.
Core materials Fuel cell configuration Pmax (mW
cm-2)
Temperature
()
Paper
number
Pr-doped ceria
(single-doped
ceria)
Ni-NCAL/Pr-doped
ceria/Ni-NCAL 744 550 II
LSCF-SCDC
composite
(in a weight
ratio 1:1)
Ni-NCAL/50LSCF-
50SCDC/Ni-NCAL
814
550
V+VI
LCPN-NCAL
composite
(in a weight
ratio 7:3)
Ni-NCAL/7LCPN-
3NCAL/Ni-NCAL
1,187
550
III
In this chapter, three different types of materials are used to investigate the fuel cell
performance based on the single-, double- and triple-doped ceria materials. Table
5.1 shows a comparison of maximum power density obtained from each of these
materials for FC devices operated at 550 . Single-element Pr-doped ceria exhibits
Doctoral thesis / Yanyan Liu
59
mixed electronic/ionic conducting properties, enabling it to be used as the core
membrane of an EFFC device with semiconducting NCAL as anode/cathode layers.
The Sm/Ca co-doped ceria is used for EFFC by combining with semiconductor
LSCF, exhibiting a comparable peak power density (814 mW cm-2 compared to
744 mW cm-2). The device using La/Pr/Nd triple-doped ceria and NCAL
nanocomposite demonstrates the highest maximum power density 1,187 mW cm-2,
an improvement of 45% or more compared to other devices.
5.5 Chapter summary
Detailed characterization studies are conducted to investigate the superficial and
structural properties of the Pr-doped CeO2 material. Additionally, the electrical
conductivity and fuel cell performance using Pr-doped CeO2 are measured to
indicate the electrochemical properties. The results exhibit that the Pr-dopant,
primary as Pr3+ ions, are selectively distributed on the surface and thus made
positive contributions to surface oxygen ion exchange and electron conduction,
which are later found to enhance fuel cell performance.
For double- and triple-doped cases, the effect of various weight ratios of ionic
conductor LCPN and semiconductor NCAL on the EFFCs performance is
investigated. The optimal sample, i.e. 7LCPN-3NCAL, exhibits a considerable fuel
cell performance with 1,187 mW cm-2 at 550 . Therefore, the electron conduction
might balance well with the ion conductivity for this LCPN/NCAL composite with
30 wt. % NCAL. The high OCV and its fast charge transfer properties demonstrate
that the mixed semiconductor (NCAL) and ionic conductor (LCPN) composite has
the good catalytic capacity to function EFFC performance well.
Doctoral thesis / Yanyan Liu
60
Doctoral thesis / Yanyan Liu
61
Chapter 6
Natural minerals for EFFC
A natural mineral originating from the natural chalcopyrite ore (written as CuFeS2)
is considered for applications in LT-SOFCs. Three types of fuel cell devices are
fabricated aiming at exploring the optimum option for the application of CuFe2O4.
EIS is employed as the principal means to reveal possible underlying working
mechanisms for the different devices. These investigations on natural minerals have
been published in Paper IV.
6.1 Material characterization analysis
Figure 6.1 displays the XRD results of untreated natural mineral and the heat-
treated samples, respectively, at 600, 700, and 900 . The main constituent of the
natural material in Figure 6.1a relates to a chalcopyrite structure (i.e. CuFeS2)
according to JCPDS NO.37-0471. Some other phases, such as FeS2, Fe2O3, CaCO3,
and SiO2, are also found in the raw mineral (as listed in Table 6.1). After heat
treatment (calcined at 600 for 2 h), the chalcopyrite structure is destroyed, see
Figure 6.1b. When elevating the temperature to 900 , some diffraction peaks are
indexed to a face-centered cubic CuFe2O4 structure (JCPDS NO. 25-0283). In this
sample, some other phases, e.g. CaSO4 and CaO, are also detected. The structural
changes for this natural mineral could be attributed to the chemical reactions during
the heating process. As reported previously, the impurity phases of the as-obtained
sample are favorable for ion transportation to improve ionic conductivity by
combining to form composites and induce interfacial conduction [156].
Figure 6.2a and 2b display the microstructures for the raw chalcopyrite ore and the
as-sintered delafossite, respectively, by means of SEM technique. As observed, the
natural chalcopyrite consists of variable bulk agglomeration in a size scale ranging
Doctoral thesis / Yanyan Liu
62
from micrometer to nanometer. After heat-treatment at 900 , the obtained
CuFe2O4 sample exhibits a slight smaller bulk structure. The quantitative analysis
of elements for this mineral is carried out by EDS measurement, indicating the
distinct Fe, Cu, and S peaks and some elemental peaks of Mg, Si, Al and Ca in the
raw chalcopyrite (see Figure 6.2c).
Figure 6.1 XRD patterns for a) raw material and b) the as-treated mineral
samples sintered at 600, 700 and 900 , respectively.
Figure 6.2 SEM images of a) raw mineral, b) the sample sintered at 900 , and
c) EDS image of raw mineral.
Doctoral thesis / Yanyan Liu
63
6.2 Electrochemical performance analysis
Figure 6.3a demonstrates the EIS results for the three samples under air
atmosphere. As observed, all of three EIS patterns display two distinct components
including a medium-frequency arc followed by a low-frequency tail. The EIS
model comparison shows that the sample obtained at 600 has much lower
resistance compared to the other two samples. It is known that CuFeS2 has the
capacity to transfer electrons, thus the undecomposed parts inside this sample
(600 ) can contribute to the entire conductivity. The middle-frequency semicircle,
generally, is indexed to the grain boundary resistance of materials. In this case, the
sample obtained at 900 exhibits a reduced grain-boundary arc, implying that the
formation of CuFe2O4 phase might facilitate to improve the conductivity of
materials. To further evaluate the electrochemical properties of the obtained sample
at 900 (mainly CuFe2O4), the EIS measurement is conducted for this sample
under fuel cell operating conditions. Figure 6.3b demonstrates an obvious reducing
EIS curve compared to the one measured in the air for the sample (900 ). This
can be attributed to the introduction of extrinsic ions, such as protons in this case,
while the electrical conduction is contributed by the intrinsic charge carriers
operated at air [104]. It is worth noting that the interfaces of this sample created by
the impurity phases are favorable for proton transportation. Therefore, the enhanced
conductivity might be caused by the proton conduction. It has been investigated
that the conductivity of an ionic conductor is improved by combining with an
insulator, such as LiFeO2 and γ-LiAlO2 [100]. In addition, the ionic conductor
LZSDC composited with the CuFe2O4-based sample is expected to enable a dual-
ion conducting (O2-/H+) function as the SOFC material, see the schematic
illustration in Figure 6.4.
Table 6.1 The main chemical composition of raw chalcopyrite.
Element MgO Al2O3 SiO2 SO3
Wt. % 6.957 3.082 21.773 28.315
Element CaO Fe2O3 CuO
Wt. % 4.247 20.840 13.959
Doctoral thesis / Yanyan Liu
64
Figure 6.3 a) Nyquist plots of a) the samples sintered at 600, 700 and 900
measured in air atmosphere at 550 ; b) the sample sintered at 900 measured
in H2/air at 550 .
Figure 6.4 Mechanism for the ionic conductivity of the LZSDC-CuFe2O4
composite.
Figure 6.5 includes schematic diagrams of the three different fuel cell devices, and
the I-V and I-P characteristics for the corresponding devices are demonstrated in
Figure 6.6a. For Type I, LZSDC and CuFe2O4 are employed as the electrolyte and
ORR catalyst, respectively. This device displays a quick-response OCV (1.07 V)
and corresponding peak power density of 180 mW cm-2 at 550 . Another device
using CuFe2O4 as both anode and cathode is fabricated, exhibiting a high OCV of
1.15 V. These findings indicate that CuFe2O4 sample has good catalytic activities
Doctoral thesis / Yanyan Liu
65
on fuel cell reactions. In Type II, the CuFe2O4 sample, composited with LZSDC, is
used as the functional cathode/anode layers, while NCAL is pasted on nickel foam
attached on both sides as the current collectors and structural supports. This device
exhibits a good power density of 354 mW cm-2 at 550 with an OCV of 0.94 V.
In contrast, the Type III device is designed to substitute LZSDC electrolyte using
the LZSDC/CuFe2O4/NCAL composite. Remarkably, the power density of Type III
is significantly enhanced up to 587 mW cm-2 at 550 , with good obtained OCV
(0.988 V). The improved fuel cell performance for Type III is attributed to the
removal of electrode/electrolyte interfaces compared to the conventional fuel cell
structure, e.g. Type I and Type II. Figure 6.6b shows an operating stability
recording for around 5 hours to preliminarily test the practical feasibility of the
Type III device. As shown, a relatively stable potential output is shown under a
steady current condition (50 mA cm-2).
Figure 6.5 Schematic illustrations for a) Type I, b) Type II and c) Type III.
To further clarify kinetic reactions, EIS measurements are carried out on the three
devices, as shown in Figure 6.7. Type I and Type II display a similar EIS model
with a half-arc at high frequency and an upturned tail at low frequency region. They
can be fitted and analyzed by means of an empirical equivalent circuit, denoted as
Ro(R1-QPE1)W1 [157]. According to these results, the high-frequency capacitance
has a level of 10-6 F for both devices. Therefore, the resistance (R1) from the
semicircle at the high frequency region can be assigned to the charge transfer
process along with the interfaces of electrodes and electrolyte [35]. The R1 values
for Type I and Type II are 0.256 Ω cm2 and 0.112 Ω cm2, respectively. The reduced
resistance loss for Type II might be caused by the appended LZSDC/CuFe2O4
functional layers. For Type III, a new EIS feature with a depressed semicircle is
displayed in Figure 6.7c. It is fitted and analyzed using the empirical circuit of
Doctoral thesis / Yanyan Liu
66
Ro(R1-QPE1)(R2-QPE2), which is employed generally by the perovskite solar cells
[35]. A capacitance with a level of 10-2-10-1 F is obtained, representing the
dominant processes for oxygen adsorption or dissociation, as well as diffusion.
Simultaneously, the polarization tail, indexing to the Warburg impedance in the
conventional fuel cell devices, is not observed for Type III. All the EIS fitting
results for the three devices are summarized in Table 6.2. By comparative analysis
on the three configurations, the Type III using the LZSDC/CuFe2O4/NCAL exhibits
much less polarization loss owing to the removal the interface resistance from
electrode and electrolyte in the conventional three-component devices. Based on
the analysis above, this device using CuFe2O4 combined with LZSDC and NCAL
as core components exhibits good ORR catalytic and ion transport properties, thus
providing good fuel cell performance.
Figure 6.6 a) I-V-P curves for three types of devices measured at 550 ; b)
Durability test of Type III at a constant discharge current of 50 mA cm-2.
Table 6.2 The fitting results of impedance spectra for three types of devices measured
in H2/air condition at 550 . The unit for resistance is Ω cm2.
Parameter Ro R1 QPE1-Q QPE1-
n
W1-
R W1-T
W1-
p R2
QPE2-
Q
QPE
2-n
Type I 0.67
8 0.4 5.3E-5 1 2.42 28 0.17 - - -
Type II 0.7 0.17
6 8.8E-5 0.8 9.5 11300 0.22 - - -
Type III 0.65
6
0.07
9 0.039 0.777 - - - 3.5 0.058 0.54
Doctoral thesis / Yanyan Liu
67
Figure 6.7 Nyquist plots for a) Type I, b) Type II and c) Type III operated at fuel
cell condition at 550 .
6.3 Chapter summary
In this study, a natural mineral material is investigated and applied successfully for
LT-SOFCs. The one using CuFe2O4 cathode exhibits a quick-response OCV,
reaching up to 1.07 V, indicating good catalytic activity of the natural mineral for
oxygen reduction. Compared to two conventional fuel cell devices, the one using a
homogeneous LZSDC/CuFe2O4/NCAL layer demonstrates a much better power
output of 587 mW cm-2 at 550 . These findings suggest that natural minerals have
great potential to be developed as new SOFC materials.
Doctoral thesis / Yanyan Liu
68
Doctoral thesis / Yanyan Liu
69
Chapter 7
Conclusions and future outlook
In this dissertation work, rare-earth doped ceria materials are employed to explore
effective approaches for desirable solid oxide fuel cell materials. Various material
synthesis techniques are investigated including two-step wet-chemical co-
precipitation method and hydrothermal method. Furthermore, a natural mineral is
investigated as an initial exploration for SOFC applications. Several conclusions
are drawn from this thesis:
i) An effective approach is developed to significantly enhance ionic
conductivity of doped ceria oxide via a novel two-step wet-chemical
co-precipitation method realizing Sm3+ doped in bulk and Pr3+/Nd3+
doped in surface domains of ceria. The modified surface property
greatly reduced the grain boundary resistance, leading to an
exceptional electrical conductivity of 0.125 S cm-1 at 600 . The
presence of Pr3+/Nd3+ also enhances the surface oxygen exchange rate
and helps to improve the redox and electrode reaction kinetics, greatly
enhancing the voltage and power efficiency. SOFC based on PNSDC
electrolyte with a thickness of 0.5 mm obtained a maximum power
density of 710 mW cm-2 at 550 . Thus, the improved electrical and
redox properties offer a new promising approach for practical
application of ceria-based materials for high performance, low
temperature SOFCs applications, which will lead to a great impact on
SOFCs development and commercialization.
ii) A single-phase Pr-doped CeO2 is synthesized via facile template-free
hydrothermal method. The detailed characterization studies are
measured to investigate the superficial and structural properties of the
Pr-doped CeO2 material. Additionally, the electrical conductivity and
fuel cell performance are conducted to indicate the electrochemical
Doctoral thesis / Yanyan Liu
70
properties. The results exhibit that the Pr dopants, primary as Pr3+ ions,
prefer to dominatingly distribute on the surface and make significant
contributions to surface properties, especially surface oxygen ion
exchange and electron conduction. Furthermore, more oxygen
vacancies and emerging electronic conduction presented on the surface
of Pr-doped CeO2 material should take responsibility for the enhanced
electrical conductivity and improved fuel cell performance in
comparison to Pr-free CeO2 sample.
iii) The rare-earth doped ceria materials are successfully applied for EFFC
devices by compositing with semiconductor materials. The effects of
the weight ratios for LCPN and NCAL on the electrochemical
performance of EFFC devices are systematically investigated. The
optimal device using 30 wt. % NCAL in the LCPN/NCAL composite
exhibits a good power output with 1,187 mW cm-2 at 550 oC. The
investigations indicate that it is the indispensable prerequisite to
acquire good fuel cell performance using a well-balanced electronic
phase and ionic conductor for EFFC devices.
iv) Natural mineral is heat-treated at 900 to obtain the sample with most
CuFe2O4 and then is applied for LT-SOFCs. Three fuel cell devices are
fabricated in different configurations. Among these results, the type
with the homogenous LZSDC/CuFe2O4/NCAL composite layer
exhibits the maximum power density reaching up to 587 mW cm-2 at
550 oC. The investigations reveal the promising properties of natural
mineral for SOFC applications, including i) high cathodic ORR
catalyst functions; ii) the potential to be served as the component of
EFFC devices with good fuel cell performance; iii) the different
electrochemical processes occurred at the natural mineral-based EFFC
device. Based on these investigations, the natural minerals including
the used CuFe2O4, exhibit great potentials to apply for next-generation
mineral-based SOFCs.
Further research work is suggested to study the following aspects:
• To develop Pr-doped ceria nanocomposite with other materials, e.g.
ionic conductors, insulators, even electronic conductors.
• To explore new materials of solid oxide fuel cell to be technically
useful at relatively low operating temperature, e.g. 400 .
Doctoral thesis / Yanyan Liu
71
• To use the tri-doped ceria for electrolyte film of SOFC to further
enhance the ionic conductivity and lower the operating temperature.
• To develop scale-up SOFC devices using the developed materials.
Doctoral thesis / Yanyan Liu
72
Doctoral thesis / Yanyan Liu
73
References
[1] X.G. Li, Taylor, Francis, Principles of fuel cells, Wiley, (2006).
doi:10.1201/9780203942338.
[2] I. Staffel, D. Scamman, A.V. Abad, P. Balcombe, P.E. Dodds, P. Ekins, N.
Shah, K.R. Ward, The role of hydrogen and fuel cells in the global energy system,
Energy Environ. Sci. 12 (2019) 463-491. doi:10.1039/c8ee01157e.
[3] M.J. Burke, J.C. Stephens, Political power and renewable energy futures: A
critical review, Energy Res. Soc. Sci. 35 (2018) 78-93.
doi:10.1016/j.erss.2017.10.018.
[4] G.W. Crabtree, M.S. Dresselhaus, M. V Buchanan, The hydrogen economy,
Physics Today 57 (2004) 39-45. doi.org/10.1063/1.1878333
[5] Z.H. Pu, J.H. Zhao, I.S. Amiinu, W.Q. Li, M. Wang, D.P. He, S.C. Mu, A
universal synthesis strategy for P-rich noble metal diphosphide-based
electrocatalysts for the hydrogen evolution reaction, Energy Environ. Sci. 12 (2019)
952-957. doi:10.1039/c9ee00197b.
[6] G. Sievi, D. Geburtig, T. Skeledzic, A. Bösmann, P. Preuster, O. Brummel, F.
Waidhas, M.A. Montero, P. Khanipour, I. Katsounaros, J. Libuda, K.J.J.
Mayrhofer, P. Wasserscheid, Towards an efficient liquid organic hydrogen carrier
fuel cell concept, Energy Environ. Sci. 12 (2019) 2305-2314.
doi:10.1039/c9ee01324e.
[7] D. Hart, J. Howes, B. Madden, E. Boyd, Hydrogen and fuel cells: opportunities
for growth a roadmap for the UK, E4tech and Element Energy, (2016).
[8] W.R. Grove, M.A. Swansea, On voltaic series and the combination of gases by
platinum, Philos. Mag. Ser. 3, (1839) 127-130. doi:10.1080/14786443908649684.
[9] J.H. Shim, Ceramics breakthrough, Nat. Energy 3 (2018) 168-169.
doi:10.1038/s41560-018-0110.
[10] N. Mahato, A. Banerjee, A. Gupta, S. Omar, K. Balani, Progress in material
selection for solid oxide fuel cell technology: A review, Prog. Mater. Sci. 72 (2015)
Doctoral thesis / Yanyan Liu
74
141-337. doi:10.1016/j.pmatsci.2015.01.001.
[11] E.D. Wachsman, K.T. Lee, Lowering the temperature of solid oxide fuel cells,
Science. 334 (2011) 935-940. doi: 10.1126/science.1204090.
[12] J.G. Barriocanal, A.R. Calzada, M. Varela, Z. Sefrioui, E. Iborra, C. Leon, S.J.
Pennycook, J. Santamaria, Colossal ionic conductivity at interfaces of epitaxial
ZrO2: Y2O3/SrTiO3 heterostructures. Science. 321 (2008) 676-681. doi:
10.1126/science.1156393.
[13] M. Li, M. Zhao, F. Li, W. Zhou, V.K. Peterson, X. Xu, Z. Shao, I. Gentle, Z.
Zhu, A niobium and tantalum co-doped perovskite cathode for solid oxide fuel
cells operating below 500 , Nat. Commun. 8 (2017) 1-9.
doi:10.1038/ncomms13990.
[14] N. Jaiswal, K. Tanwar, R. Suman, D. Kumar, S. Upadhyay, O. Parkash, A
brief review on ceria based solid electrolytes for solid oxide fuel cells, J. Alloys
Compd. 781 (2019) 984-1005. doi:10.1016/j.jallcom.2018.12.015.
[15] J.B. Goodenough, Ceramic technology: oxide-ion conductors by design,
Nature 404 (2000) 821-823. doi:10.1038/35009177.
[16] B.C.H. Steele, A. Heinzel, Materials for fuel-cell technologies, Nature (2010)
345-352. doi:10.1142/9789814317665_0031.
[17] E. Fabbri, L. Bi, D. Pergolesi, E. Traversa, Towards the next generation of
solid oxide fuel cells operating below 600 with chemically stable proton-
conducting electrolytes, Adv. Mater. (2012) 195-208.
doi:10.1002/adma.201103102.
[18] R.V. Basshuysen, F. Schäfer, Internal combustion engine handbook: basics,
components, systems, and perspectives, Wiesbaden. (2004).
[19] National Renewable Energy Laboratory, 1-10 kW Stationary combined heat
and power systems status and technical potential, Independent review, (2010).
[20] S.C. Singhal, K. Kendall, High temperature solid oxide fuel cells:
Fundamentals, design and applications, 1st Ed. Elesevier, (2003).
[21] B.A. Boukamp, Fuel cells: the amazing perovskite anode, Nat. Mater. 2 (2003)
294-296. doi:10.1038/nmat892.
[22] Y.L. Huang, A.M. Hussain, E.D. Wachsman, Nanoscale cathode modification
for high performance and stable lowtemperature solid oxide fuel cells (SOFCs),
Nano Energy 49 (2018) 186-192. doi: 10.1016/j.nanoen.2018.04.028.
Doctoral thesis / Yanyan Liu
75
[23] D.J.L. Brett, A. Atkinson, P. Brandon, S.J. Skinner, A. Atkinson, D.J.L. Brett,
Intermediate temperateure solid oxide fuel cells, (2008) 1568-1578.
doi:10.1039/b612060c.
[24] K. Huang, R.S. Tichy, J.B. Goodenough, Phase relationships and electrical
properties, 75 (1998).
[25] T. Ishihara, H. Matsuda, Y. Takita, Doped LaGaOg perovskite type oxide as
conductor, (1994) 3801-3803. doi:10.1021/ja00088a016.
[26] M. Mogensen, S. Skaarup, Kinetic and geometric aspects of solid oxide fuel
cell electrodes, Solid State Ionics. 86 (1996) 1151-1160. doi:10.1016/0167-
2738(96)00280-9.
[27] X. Chen, N.J. Wu, L. Smith, A. Ignatiev, Thin-film heterostructure solid oxide
fuel cells, Appl. Phys. Lett. 84 (2016) 1-4. doi:10.1063/1.1697623.
[28] S.D. Souza, S.J. Visco, L.C. De Jonghe, Thin-film solid oxide fuel cell with
high performance at low-temperature, Solid State Ionics. 98 (1997) 57-61.
[29] X. Guo, Comment on 'colossal ionic conductivity at interfaces of epitaxial
ZrO2:Y2O3/SrTiO3 heterostructures', Nature. 324 (2009) 3-5.
[30] S.W. Tao, J.T.S. Irvine, A stable, easily sintered proton-conducting oxide
electrolyte for moderate-temperature fuel cells and electrolyzers, Adv. Mater. 18
(2006) 1581-1584. doi:10.1002/adma.200502098.
[31] B. Zhu, R. Raza, G. Abbas, M. Singh, An electrolyte-free fuel cell constructed
from one homogenous layer with mixed conductivity, Adv. Funct. Mater. 21 (2011)
2465-2469. doi:10.1002/adfm.201002471.
[32] B. Zhu, R. Raza, H. Qin, Q. Liu, L. Fan, Fuel cells based on electrolyte and
non-electrolyte separators, Energy Environ. Sci. 4 (2011) 2986-2992.
doi:10.1039/c1ee01202a.
[33] B. Zhu, L. Fan, P. Lund, Breakthrough fuel cell technology using ceria-based
multi-functional nanocomposites, Appl. Energy. 106 (2013) 163-175.
doi:10.1016/j.apenergy.2013.01.014.
[34] B. Zhu, P.D. Lund, R. Raza, Y. Ma, L. Fan, M. Afzal, J. Patakangas, Y. He,
Y. Zhao, W. Tan, Q.A. Huang, J. Zhang, H. Wang, Schottky junction effect on
high performance fuel cells based on nanocomposite materials, Adv. Energy Mater.
5 (2015) 1-6. doi:10.1002/aenm.201401895.
[35] B. Zhu, Y. Huang, L. Fan, Y. Ma, B. Wang, C. Xia, M. Afzal, B. Zhang, W.
Doctoral thesis / Yanyan Liu
76
Dong, H. Wang, P.D. Lund, Novel fuel cell with nanocomposite functional layer
designed by perovskite solar cell principle, Nano Energy. 19 (2016) 156-164.
doi:10.1016/j.nanoen.2015.11.015.
[36] B. Zhu, B. Wang, Y. Wang, R. Raza, W. Tan, J.S. Kim, P.A. van Aken, P.
Lund, Charge separation and transport in La0.6Sr0.4Co0.2Fe0.8O3-δ and ion-doping
ceria heterostructure material for new generation fuel cell, Nano Energy. 37 (2017)
195-202. doi:10.1016/j.nanoen.2017.05.003.
[37] Y. Wu, C. Xia, W. Zhang, X. Yang, Z.Y. Bao, J.J. Li, B. Zhu, Natural hematite
for next-generation solid oxide fuel cells, Adv. Funct. Mater. 26 (2016) 938-942.
doi:10.1002/adfm.201503756.
[38] P.D. Lund, B. Zhu, Y. Li, S. Yun, A.G. Nasibulin, R. Raza, M. Leskelä, M.
Ni, Y. Wu, G. Chen, L. Fan, J.-S. Kim, S. Basu, T. Kallio, I. Pamuk, Standardized
procedures important for improving single-component ceramic fuel cell
technology, ACS Energy Lett. (2017) 2752-2755.
doi:10.1021/acsenergylett.7b00997.
[39] B. Zhu, Y. Huang, L. Fan, Y. Ma, B. Wang, C. Xia, M. Afzal, B. Zhang, W.
Dong, H. Wang, P.D. Lund, Novel fuel cell with nanocomposite functional layer
designed by perovskite solar cell principle, Nano Energy. 19 (2016) 156-164.
doi:10.1016/j.nanoen.2015.11.015.
[40] H. Hu, Q. Lin, Z. Zhu, B. Zhu, X. Liu, Fabrication of electrolyte-free fuel cell
with Mg0.4Zn0.6O/Ce0.8Sm0.2O2-δ-Li0.3Ni0.6Cu0.07Sr0.03O2-δ layer, J. Power Sources.
248 (2014) 577-81. doi:10.1016/j.jpowsour.2013.09.095.
[41] B. Zhu, P. Lund, R. Raza, J. Patakangas, Q.A. Huang, L. Fan, M. Singh, A
new energy conversion technology based on nano-redox and nano-device
processes, Nano Energy. 2 (2013) 1179-1185. doi:10.1016/j.nanoen.2013.05.001.
[42] Y. Chen, B.D. Glee, Y. Tang, Z.Y. Wang, B.T. Zhao, Y.C. Wei, L. Zhang, S.
Yoo, K. Pei, J.H. Kim, Y. Ding, P. Hu, F.F. Tao, M.L. Liu, A robust fuel cell
operated on nearly dry methane at 500 °C enabled by synergistic thermal catalysis
and electrocatalysis, Nature. 3 (2018). doi:10.1038/s41560-018-0262-5.
[43] C.W. Sun, H. Li, L.Q. Chen, Nanostructured ceria-based materials: synthesis,
properties, and applications, Energy Environ. Sci. 5 (2012) 8475-8505.
doi:10.1039/c2ee22310d.
[44] D. Marrocchelli, S.R. Bishop, H.L. Tuller, B. Yildiz, Understanding Chemical
expansion in non-stoichiometric oxides: ceria and zirconia case studies, (2012)
Doctoral thesis / Yanyan Liu
77
1958-1965. doi:10.1002/adfm.201102648.
[45] B.S. Sanna, V. Esposito, D. Pergolesi, A. Orsini, A. Tebano, S. Licoccia, G.
Balestrino, E. Traversa, Fabrication and electrochemical properties of epitaxial
samarium-doped ceria films on SrTiO3-buffered MgO substrates, Adv. Funct.
Mater. (2009) 1713-1719. doi:10.1002/adfm.200801768.
[46] W. Lee, H.J. Jung, M.H. Lee, Y. Kim, J.S. Park, R. Sinclair, F.B. Prinz,
Oxygen surface exchange at grain boundaries of oxide ion conductors, Adv. Funct.
Mater. 22 (2012) 965-971. doi:10.1002/adfm.201101996.
[47] X. Ge, S. Chan, Q. Liu, Q. Sun, Solid oxide fuel cell anode materials for direct
hydrocarbon utilization, Adv. Energy Mater. (2012) 1156-1181.
doi:10.1002/aenm.201200342.
[48] A.V. Virkar, Theoretical analysis of solid oxide fuel cells with two-layer,
composite electrolytes: electrolyte stability, J. Electrochem. Soc., 138 (1991)
1481-1487.
[49] J. Hou, L. Bi, J. Qian, Z.W. Zhu, J.Y. Zhang, W. Liu, High performance ceria-
bismuth bilayer electrolyte low temperature solid oxide fuel cells (LT-SOFCS),
fabricated by combing co-pressing with drop-coating, J. Mater. Chem. A 3 (2015)
10219-10224. doi:10.1039/c4ta06864e.
[50] S. Sanna, V. Esposito, J.W. Andreasen, J. Hjelm, W. Zhang, T. Kasama, S.B.
Simonsen, M. Christensen, S. Linderoth, N. Pryds, Enhancement of the chemical
stability in confined δ-Bi2O3, Nat. Mater.14 (2015) 1-5. doi:10.1038/NMAT4266.
[51] L. Fan, B. Zhu, P. Su, Nanomaterials and technologies for low temperature
solid oxide fuel cells: recent advances, challenges and opportunities, Nano Energy.
45 (2018) 148-176. doi:10.1016/j.nanoen.2017.12.044.
[52] G. Zhang, W. Li, W. Huang, Z. Cao, K. Shao, F. Li, C. Tang, C. Li, C. He, Q.
Zhang, L. Fan, solid oxide fuel cells: one-step synthesis and super interfacial
proton conduction, J. Power Sources. 386 (2018) 56-65.
doi:10.1016/j.jpowsour.2018.03.035.
[53] S. Banerjee, P.S. Devi, D. Topwal, S. Mandal, K. Menon, Enhanced ionic
conductivity in Ce0.8Sm0.2O1.9: unique effect of calcium Co-doping, Adv. Funct.
Mater. 17 (2007) 2847-2854. doi:10.1002/adfm.200600890.
[54] A.B. Garg, A.K. Mishra, K.K. Pandey, S.M. Sharma, Multiferroic CuCrO2
under high pressure: in situ X-ray diffraction and Raman spectroscopic studies, J.
Appl. Phys. 116 (2014) 133514. doi:10.1063/1.4896952.
Doctoral thesis / Yanyan Liu
78
[55] M.S. Miao, S. Yarbro, P.T. Barton, R. Seshadri, Electron affinities and
ionization energies of Cu and Ag delafossite compounds: a hybrid functional study,
Phys. Rev. B 89 (2014) 1-8. doi:10.1103/PhysRevB.89.045306.
[56] L.O. Arteta, A.T. Aguayo, A. Remiro, J. Bilbao, A.G. Gayubo, Behavior of a
CuFe2O4/γ-Al2O3 catalyst for the steam reforming of dimethyl ether in reaction-
regeneration cycles, Ind. Eng. Chem. Res. 54 (2015) 11285-11294.
doi:10.1021/acs.iecr.5b02901.
[57] Y. Li, J. Shen, Y. Hu, S. Qiu, G. Min, Z. Song, Z. Sun, C. Li, General flame
approach to chainlike MFe2O4 spinel (M = Cu, Ni, Co, Zn) nanoaggregates for
reduction of nitroaromatic compounds, Ind. Eng. Chem. Res. 54 (2015) 9750-9757.
doi:10.1021/acs.iecr.5b02090.
[58] S. Narjes, F. Karimzadeh, M. Hossein, N. Mark, Oxidation and electrical
behavior of CuFe2O4 spinel coated crofer 22 APU stainless steel for SOFC
interconnect application, Solid State Ionics. 289 (2016) 95-105.
doi:10.1016/j.ssi.2016.02.015.
[59] W.Y. Zhang, B. Hua, N.Q. Duan, J. Pu, B. Chi, J. Li, Cu-Fe spinel coating as
oxidation barrier for Fe-16Cr metallic interconnect in solid oxide fuel cells, J.
Electrochem. Soc. 159 (2012) 388-392. doi:10.1149/2.021209jes.
[60] I. Pilatowsky, R.J. Romero, C.A. Isaza, S.A. Gamboa, P.J. Sebastian, W.
Rivera, Thermodynamics of fuel cells, Springer, London, (2011).
doi:10.1007/978-1-84996-028-1_2.
[61] D.J.L. Brett, M. Manage, E. Agante, N.P. Brandon, E. Brightman, R.J.C.
Brown, I. Staffell, Fuels and fuel processing for low temperature fuel cells, Polym.
Electrolyte Membr. Direct Methanol Fuel Cell Technol. 1 (2012) 3-26.
doi:10.1533/9780857095473.1.3.
[62] J. Larminie, A. Dicks, Fuel cell systems explained, 2nd Ed. Wiley, (2001).
doi:10.1016/S0378-7753(00)00571-1.
[63] S.P. Jiang, Y.S. Yan, Materials for high-temperature fuel cells, Wiley (2013).
doi: 10.1002/9783527644261.
[64] A. Atkinson, S. Barrnett, R.J. Gorte, J.T.S. Irvine, A.J. Mcevoy, M. Mogensen,
S.C. Singhal, J. Vohs, Advanced anodes for high-temperature fuel cells, Mater.
Sustain. Energy. (2010) 213-223. doi:10.1038/nmat1040.
[65] R.J. Kee, H. Zhu, D.G. Goodwin, Solid-oxide fuel cells with hydrocarbon
fuels, Proc. Combust. Inst. 30 II (2005) 2379-2404.
Doctoral thesis / Yanyan Liu
79
doi:10.1016/j.proci.2004.08.277.
[66] S. Tao, J.T.S. Irvine, A redox-stable efficient anode for solid-oxide fuel cells,
Nat. Mater. 2 (2003) 320-323. doi:10.1038/nmat871.
[67] C. Yang, J. Li, Y. Lin, J. Liu, F. Chen, M. Liu, In situ fabrication of CoFe
alloy nanoparticles structured (Pr0.4Sr0.6)3(Fe0.85Nb0.15)2O7 ceramic anode for direct
hydrocarbon solid oxide fuel cells, Nano Energy. 11 (2015) 704-710.
doi:10.1016/j.nanoen.2014.12.001.
[68] M. Mogensen, K.V. Jensen, M.J. Jorgensen, S. Primdahl, Progress in
understanding SOFC electrodes, Solid State Ionics. 150 (2002) 123-129.
doi:10.1016/S0167-2738(02)00269-2.
[69] N.C. Triantafyllopoulos, S.G. Neophytides, The nature and binding strength
of carbon adspecies formed during the equilibrium dissociative adsorption of
CH4on Ni-YSZ cermet catalysts, J. Catal. 217 (2003) 324-333.
doi:10.1016/S0021-9517(03)00062-9.
[70] C.M. Finnerty, N.J. Coe, R.H. Cunningham, R.M. Ormerod, Carbon
formation on and deactivation of nickel-based/zirconia anodes in solid oxide fuel
cells running on methane, Catal. Today. 46 (1998) 137-145.
[71] D. Sarantaridis, A. Atkinson, Redox cycling of Ni-based solid oxide fuel cell
anodes: A review, Fuel Cells. 7 (2007) 246-258. doi:10.1002/fuce.200600028.
[72] E. Perry Murray, T. Tsai, S.A. Barnett, A direct-methane fuel cell with a ceria-
based anode, Nature. 400 (1999) 649-651. doi:10.1038/23220.
[73] J.C. Ruiz-Morales, J. Canales-Vázquez, C. Savaniu, D. Marrero-López, W.
Zhou, J.T.S. Irvine, Disruption of extended defects in solid oxide fuel cell anodes
for methane oxidation, Nature. 439 (2006) 568-571. doi:10.1038/nature04438.
[74] B.A. Boukamp, The Amazing Perovskite Anode, Nat. Mater. 2 (2003) 294-
296.
[75] S. Sengodan, S. Choi, A. Jun, T.H. Shin, Y.W. Ju, H.Y. Jeong, J. Shin, J.T.S.
Irvine, G. Kim, Layered oxygen-deficient double perovskite as an efficient and
stable anode for direct hydrocarbon solid oxide fuel cells, Nat. Mater. 14 (2015)
205-209. doi:10.1038/nmat4166.
[76] L.M. Kolchina, N. V Lyskov, D.I. Petukhov, G.N. Mazo, Cathodes for solid
oxide fuel cells, J. Alloys Compd. 605 (2014) 89-95.
doi:10.1016/j.jpowsour.2005.08.047.
Doctoral thesis / Yanyan Liu
80
[77] Z. Gao, L. V. Mogni, E.C. Miller, J.G. Railsback, S.A. Barnett, A perspective
on low-temperature solid oxide fuel cells, Energy Environ. Sci. 9 (2016) 1602-
1644. doi:10.1039/c5ee03858h.
[78] Z. Shao, S.M. Halle, A high-performance cathode for the next generation of
solid-oxide fuel cells, Nature. 431 (2004) 170-173. doi:10.1038/nature02863.
[79] D. Ding, X. Li, S.Y. Lai, K. Gerdes, M. Liu, Enhancing SOFC cathode
performance by surface modification through infiltration, Energy Environ. Sci. 7
(2014) 552-575. doi:10.1039/c3ee42926a.
[80] K.T. Lee, A.A. Lidie, H.S. Yoon, E.D. Wachsman, Rational design of lower-
temperature solid oxide fuel cell cathodes via nanotailoring of co-assembled
composite structures, Angew. Chemie - Int. Ed. 53 (2014) 13463-13467.
doi:10.1002/anie.201408210.
[81] J. Hou, L. Miao, J. Hui, L. Bi, W. Liu, J.T.S. Irvine, A novel: in situ diffusion
strategy to fabricate high performance cathodes for low temperature proton-
conducting solid oxide fuel cells, J. Mater. Chem. A. 6 (2018) 10411-10420.
doi:10.1039/c8ta00859k.
[82] T. Takahashi, H. Iwahara, Oxide ion conductors based on bismuthsesquioxide,
Mater. Res. Bull. 13 (1978) 135-139. doi: 10.1016/0025-5408(78)90138-1.
[83] T. Ishihara, H. Matsuda, Y. Takita, Doped LaGaO3 Perovskite Type Oxide as
a New Oxide Ionic Conductor, J. Am. Chem. Soc. 116 (1994) 3801-3803.
doi:10.1021/ja00088a016.
[84] F. Abraham, J.C. Boivin, G. Mairesse, G. Nowogrocki, The bimevox series:
A new family of high performances oxide ion conductors, Solid State Ionics. 40-
41 (1990) 934-937. doi:10.1016/0167-2738(90)90157-M.
[85] P. Lacorre, F. Goutenoire, O. Bohnke, R. Retoux, Y. Lallgant, Designing fast
oxide-ion conductors based on La2Mo2O9, Nature. 404 (2000) 856-858.
doi:10.1038/35009069.
[86] T. Ishihara, Oxide ion-conducting materials for electrolytes, Mater. high-
temperature fuel cells. (2012) 97-132. doi:10.1002/9783527644261.ch3.
[87] K.D. Kreuer, Proton-conducting oxides, Annu. Rev. Mater. Res. 33 (2003)
333-359. doi:10.1146/annurev.matsci.33.022802.091825.
[88] L. Yang, Z. Liu, S. Wang, Y. Choi, C. Zuo, M. Liu, A mixed proton, oxygen
ion, and electron conducting cathode for SOFCs based on oxide proton conductors,
Doctoral thesis / Yanyan Liu
81
J. Power Sources. 195 (2010) 471-474. doi:10.1016/j.jpowsour.2009.07.057.
[89] D. Pergolesi, E. Fabbri, A. D’Epifanio, E. Di Bartolomeo, A. Tebano, S.
Sanna, S. Licoccia, G. Balestrino, E. Traversa, High proton conduction in grain-
boundary-free yttrium-doped barium zirconate films grown by pulsed laser
deposition, Nat. Mater. 9 (2010) 846-852. doi:10.1038/nmat2837.
[90] E. Fabbri, D. Pergolesi, E. Traversa, Materials challenges toward proton-
conducting oxide fuel cells: A critical review, Chem. Soc. Rev. 39 (2010) 4355-
4369. doi:10.1039/b902343g.
[91] S. Choi, C.J. Kucharczyk, Y. Liang, X. Zhang, I. Takeuchi, H. Il Ji, S.M. Haile,
Exceptional power density and stability at intermediate temperatures in protonic
ceramic fuel cells, Nat. Energy. 3 (2018) 202-210. doi:10.1038/s41560-017-0085-
9.
[92] J. Di, M. Chen, C. Wang, J. Zheng, L. Fan, B. Zhu, Samarium doped ceria-
(Li/Na)2CO3 composite electrolyte and its electrochemical properties in low
temperature solid oxide fuel cell, J. Power Sources. 195 (2010) 4695-4699.
doi:10.1016/j.jpowsour.2010.02.066.
[93] L. Fan, C. Wang, M. Chen, B. Zhu, Recent development of ceria-based
(nano)composite materials for low temperature ceramic fuel cells and electrolyte-
free fuel cells, J. Power Sources. 234 (2013) 154-174.
doi:10.1016/j.jpowsour.2013.01.138.
[94] X. Wang, Y. Ma, R. Raza, M. Muhammed, B. Zhu, Novel core-shell
SDC/amorphous Na2CO3 nanocomposite electrolyte for low-temperature SOFCs,
Electrochem. Commun. 10 (2008) 1617-1620. doi:10.1016/j.elecom.2008.08.023.
[95] B. Zhu, Functional ceria-salt-composite materials for advanced ITSOFC
applications, J. Power Sources. 114 (2003) 1-9. doi:10.1016/S0378-
7753(02)00592-X.
[96] Y. Ma, X. Wang, S. Li, M.S. Toprak, B. Zhu, M. Muhammed, Samarium-
doped ceria nanowires: Novel synthesis and application in low-temperature solid
oxide fuel cells, Adv. Mater. 22 (2010) 1640-1644. doi:10.1002/adma.200903402.
[97] L. Fan, B. Zhu, P.C. Su, C. He, Nanomaterials and technologies for low
temperature solid oxide fuel cells: recent advances, challenges and opportunities,
Nano Energy. 45 (2018) 148-176. doi:10.1016/j.nanoen.2017.12.044.
[98] J.S. J. Garcia-Barriocanal, A. Rivera-Calzada, M. Varela, Z. Sefrioui, E.
Iborra, C. Leon, S. J. Pennycook, Colossal ionic conductivity at interfaces of
Doctoral thesis / Yanyan Liu
82
epitaxial ZrO2:Y2O3/SrTiO3 Heterostructures, Science 321 (2008) 676-681.
[99] R. Chockalingam, V.R.W. Amarakoon, H. Giesche, Alumina/cerium oxide
nano-composite electrolyte for solid oxide fuel cell applications, J. Eur. Ceram.
Soc. 28 (2008) 959-963. doi:10.1016/j.jeurceramsoc.2007.09.031.
[100] R. Lan, S. Tao, High Ionic Conductivity in a LiFeO2-LiAlO2 composite
under H2/air fuel cell conditions, Chem. A Eur. J. 21 (2015) 1350-1358.
doi:10.1002/chem.201404476.
[101] B. Zhu, Y. Huang, L. Fan, Y. Ma, B. Wang, C. Xia, M. Afzal, B. Zhang, W.
Dong, H. Wang, P.D. Lund, Novel fuel cell with nanocomposite functional layer
designed by perovskite solar cell principle, Nano Energy. 19 (2016) 156-164.
doi:10.1016/j.nanoen.2015.11.015.
[102] B. Zhu, P. Lund, R. Raza, J. Patakangas, Q. Huang, L. Fan, M. Singh, A new
energy conversion technology based on nano-redox and nano-device processes,
Nano Energy. 2 (2013) 1179-1185. doi:10.1016/j.nanoen.2013.05.001.
[103] D.A. Links, A new energy conversion technology joining electrochemical
and physical, RSC Advances (2012) 5066-5070. doi:10.1039/c2ra01234k.
[104] L. Fan, Y. Ma, X. Wang, M. Singh, B. Zhu, Understanding the
electrochemical mechanism of the core-shell ceria-LiZnO nanocomposite in a low
temperature solid oxide fuel cell, J. Mater. Chem. A. 2 (2014) 5399-5407.
doi:10.1039/c3ta14098a.
[105] T. Mavri, M. Ben, R. Imani, I. Junkar, M. Valant, V. Kralj-Lgli,
Electrochemical Biosensor based on TiO2 nanomaterials for cancer diagnostics,
Elesevier, 27 (2018). doi:10.1016/bs.abl.2017.12.003.
[106] B.Y.R.D. Shannon, M. H, N.H. Baur, O.H. Gibbs, M. Eu, V. Cu, Revised
effective ionic radii and systematic studies of interatomie distances in halides and
chaleogenides central research and development department, experimental station ,
E . L Du Pont de Nemours the effective ionic radii of Shannon & Prewitt (1976).
[107] U. Holzwarth, N. Gibson, The Scherrer equation versus the Debye-Scherrer
equation., Nat. Nanotechnol. 6 (2011) 534. doi:10.1038/nnano.2011.145.
[108] M. Guo, J.Q. Lu, Y.N. Wu, Y.J. Wang, M.F. Luo, UV and visible raman
studies of oxygen vacancies in rare-earth-doped ceria, Langmuir, 27 (2011) 3872-
3877. doi:10.1021/la200292f.
[109] X. Yang, L. Yang, S. Lin, R. Zhou, New insight into the doping effect of
Doctoral thesis / Yanyan Liu
83
Pr2O3 on the structure-activity relationship of Pd/CeO2-ZrO2 catalysts by Raman
and XRD rietveld analysis, J. Phys. Chem. C. 119 (2015) 6065-6074.
doi:10.1021/jp512606m.
[110] W. Gao, Z. Zhang, J. Li, Y. Ma, Y. Qu, Surface engineering on CeO2
nanorods by chemical redox etching and their enhanced catalytic activity for CO
oxidation, Nanoscale. 7 (2015) 11686-11691. doi:10.1039/C5NR01846C.
[111] R.C. Maher, P.R. Shearing, E. Brightman, D.J.L. Brett, N.P. Brandon, L.F.
Cohen, Reduction dynamics of doped ceria, nickel oxide, and cermet composites
probed using in situ Raman spectroscopy, Adv. Sci. (2015).
doi:10.1002/advs.201500146.
[112] I. Kosacki, T. Suzuki, H.U. Anderson, P. Colomban, Raman scattering and
lattice defects in nanocrystalline CeO2 thin films, Solid State Ionics. 149 (2002)
99-105. doi:10.1016/S0167-2738(02)00104-2.
[113] S. Wang, W. Wang, J. Zuo, Y. Qian, Study of the Raman spectrum of CeO2
nanometer thin films, Mater. Chem. Phys. 68 (2001) 246-248. doi:10.1016/S0254-
0584(00)00357-6.
[114] S. Babu, R. Thanneeru, T. Inerbaev, R. Day, A.E. Masunov, A. Schulte, S.
Seal, Dopant-mediated oxygen vacancy tuning in ceria nanoparticles.,
Nanotechnology. 20 (2009) 085713. doi:10.1088/0957-4484/20/8/085713.
[115] M. Guo, J.Q. Lu, Y.N. Wu, Y.J. Wang, M.F. Luo, UV and visible raman
studies of oxygen vacancies in rare-earth-doped ceria, Langmuir, 27 (2011) 3872-
3877. doi:10.1021/la200292f.
[116] Y.M. and Y.Q. Wei Gao, Zhiyun Zhang, Jing Li, Surface engineering on
CeO2 nanorods by chemical redox etching and their enhanced catalytic activity for
CO oxidation, Nanoscale. 7 (2015) 11686-11691. doi:10.1039/c5nr01846c.
[117] F. Jin, Y. Shen, R. Wang, T. He, Double-perovskite PrBaCo2/3Fe2/3Cu2/3O5+δ
as cathode material for intermediate- temperature solid-oxide fuel cells, J. Power
Sources. 234 (2013) 244-251. doi:10.1016/j.jpowsour.2013.01.172.
[118] D. He, H. Hao, D. Chen, J. Liu, J. Yu, J. Lu, F. Liu, G. Wan, S. He, Y. Luo,
Synthesis and application of rare-earth elements (Gd, Sm, and Nd) doped ceria-
based solid solutions for methyl mercaptan catalytic decomposition, Catal. Today.
281 (2017) 559-565. doi:10.1016/j.cattod.2016.06.022.
[119] S. Deshpande, S. Patil, S.V. Kuchibhatla, S. Seal, Size dependency variation
in lattice parameter and valency states in nanocrystalline cerium oxide, Appl. Phys.
Doctoral thesis / Yanyan Liu
84
Lett. 87 (2005) 1-3. doi:10.1063/1.2061873.
[120] H. Borchert, Y. Borchert, V. V. Kaichev, I.P. Prosvirin, G.M. Alikina, A.I.
Lukashevich, V.I. Zaikovskii, E.M. Moroz, E.A. Paukshtis, V.I. Bukhtiyarov, V.A.
Sadykov, Nanostructured, Gd-doped ceria promoted by Pt or Pd: Investigation of
the electronic and surface structures and relations to chemical properties, J. Phys.
Chem. B. 109 (2005) 20077-20086. doi:10.1021/jp051525m.
[121] M. Piumetti, S. Bensaid, N. Russo, D. Fino, Nanostructured ceria-based
catalysts for soot combustion: Investigations on the surface sensitivity, Appl. Catal.
B Environ. 165 (2015) 742-751. doi:10.1016/j.apcatb.2014.10.062.
[122] Q. Shen, M. Wu, H. Wang, C. He, Z. Hao, W. Wei, Y. Sun, Facile synthesis
of catalytically active CeO2 for soot combustion, Catal. Sci. Technol. (2015) 1941-
1952. doi:10.1039/C4CY01435A.
[123] F. Jin, H. Xu, W. Long, Y. Shen, T. He, Characterization and evaluation of
double perovskites LnBaCoFeO5+δ (Ln = Pr and Nd) as intermediate-temperature
solid oxide fuel cell cathodes, J. Power Sources. 243 (2013) 10-18.
doi:http://dx.doi.org/10.1016/j.jpowsour.2013.05.187.
[124] F. Jin, Y. Shen, R. Wang, T. He, Double-perovskite PrBaCo2/3Fe2/3Cu2/3O5
as cathode material for intermediatetemperature solid-oxide fuel cells, J. Power
Sources. 234 (2013) 244-251. doi:10.1016/j.jpowsour.2013.01.172.
[125] S. Kuharuangrong, Ionic conductivity of Sm, Gd, Dy and Er-doped ceria, J.
Power Sources. 171 (2007) 506-510. doi:10.1016/j.jpowsour.2007.05.104.
[126] B.J.T.S. Irvine, D.C. Sinclair, A.R. West, Electroceramics: characterization
by impedance spectroscopy, 2 (1990) 132-138.
[127] Q. Li, T. Xia, X.D. Liu, X.F. Ma, J. Meng, X.Q. Cao, Fast densification and
electrical conductivity of yttria-stabilized zirconia nanoceramics, 138 (2007) 78-
83. doi:10.1016/j.mseb.2006.12.012.
[128] T.S. Oh, S.M. Haile, Electrochemical behavior of thin-film Sm-doped ceria:
insights from the point-contact configuration, Phys. Chem. Chem. Phys. 17 (2015)
13501-13511. doi:10.1039/C4CP05990E.
[129] L. Fan, H. Zhang, M. Chen, C. Wang, H. Wang, M. Singh, B. Zhu,
Electrochemical study of lithiated transition metal oxide composite as symmetrical
electrode for low temperature ceramic fuel cells, Int. J. Hydrogen Energy. 38 (2013)
11398-11405. doi:10.1016/j.ijhydene.2013.06.050.
Doctoral thesis / Yanyan Liu
85
[130] S.B. Adler, Factors governing oxygen reduction in solid oxide fuel cell
cathodes, Chem. Rev. 104 (2004) 4791-4843. doi:Doi 10.1021/Cr020724o.
[131] H. He, H.X. Dai, C.T. Au, Defective structure, oxygen mobility, oxygen
storage capacity, and redox properties of RE-based (RE=Ce, Pr) solid solutions,
Catal. Today. 90 (2004) 245-254. doi:10.1016/j.cattod.2004.04.033.
[132] M.Y. Sinev, Kinetic and structural studies of oxygen avaiablity of the mixed
oxides Pr1-xMxOy (M=Ce , Zr), J. Mater. Res. 8 (1996).
[133] H. Shimada, T. Yamaguchi, T. Suzuki, H. Sumi, K. Hamamoto, Y. Fujishiro,
High power density cell using nanostructured Sr-doped SmCoO3 and Sm-doped
CeO2 composite powder synthesized by spray pyrolysis, J. Power Sources. 302
(2016) 308-314. doi:10.1016/j.jpowsour.2015.10.082.
[134] M. Nolan, J.E. Fearon, G.W. Watson, Oxygen vacancy formation and
migration in ceria, Solid State Ionics. 177 (2006) 3069-3074.
doi:10.1016/j.ssi.2006.07.045.
[135] L. Fan, B. Zhu, M. Chen, C. Wang, R. Raza, H. Qin, X. Wang, X. Wang, Y.
Ma, High performance transition metal oxide composite cathode for low
temperature solid oxide fuel cells, J. Power Sources. 203 (2012) 65-71.
doi:10.1016/j.jpowsour.2011.12.017.
[136] N. Dai, Z. Lou, Z. Wang, X. Liu, Y. Yan, J. Qiao, T. Jiang, K. Sun, Synthesis
and electrochemical characterization of Sr2Fe1.5Mo0.5O6-Sm0.2Ce0.8O1.9 composite
cathode for intermediate-temperature solid oxide fuel cells, J. Power Sources. 243
(2013) 766-772. doi:10.1016/j.jpowsour.2013.05.168.
[137] T. Wei, Q. Zhang, Y.-H. Huang, J.B. Goodenough, Cobalt-based double-
perovskite symmetrical electrodes with low thermal expansion for solid oxide fuel
cells, J. Mater. Chem. 22 (2012) 225. doi:10.1039/c1jm14756k.
[138] B.C. Steele, A. Heinzel, Materials for fuel-cell technologies., Nature. 414
(2001) 345-352. doi:10.1038/35104620.
[139] K. Ahn, D.S. Yoo, D.H. Prasad, H. Lee, Y. Chung, J. Lee, Role of
multivalent Pr in the formation and migration of oxygen vacancy in Pr-doped ceria:
experimental and first-principles investigations, Chem. Mater. 24 (2012) 4261-
4267. doi:10.1021/cm3022424.
[140] Y. Chen, Y. Chen, D. Ding, Y. Ding, Y.M. Choi, L. Zhang, S. Yoo, D. Chen,
B. DeGlee, H. Xu, Q. Lu, B. Zhao, G. Vardar, J. Wang, H. Bluhm, E.J. Crumlin,
C. Yang, J. Liu, B. Yildiz, M. Liu, A robust and active hybrid catalyst for facile
Doctoral thesis / Yanyan Liu
86
oxygen reduction in solid oxide fuel cells, Energy Environ. Sci. 10 (2017) 964-
971. doi:10.1039/c6ee03656b.
[141] D.N. Durgasri, T. Vinodkumar, P. Sudarsanam, B.M. Reddy, Nanosized
CeO2-Gd2O3 mixed oxides: Study of structural characterization and catalytic CO
oxidation activity, Catal. Letters. 144 (2014) 971-979. doi:10.1007/s10562-014-
1223-7.
[142] W. Gao, Z.Y. Zhang, J. Li, Y.Y. Qu, Surface engineering on CeO2 nanorods
by chemical redox etching and their enhanced catalytic activity for CO oxidation,
Nanoscale 7 (2015) 11686-11691. doi:10.1039/c5nr01846c.
[143] M.F. Luo, Z.L. Yan, L.Y. Jin, M. He, Raman spectroscopic study on the
structure in the surface and the bulk shell of CexPr1-xO2-δ mixed oxides, J. Phys.
Chem. B 26 (2006) 13068-13071. doi:10.1021/jp057274z.
[144] Z.X. Zhang, Y.H. Wang, J.M. Lu, J. Zhang, M.R. Li, X.B. Liu, F. Wang, Pr-
doped CeO2 catalyst in the prins condensation-hydrolysis reaction: are all of the
defect sites catalytically active? ACS. Catal. 8 (2018) 2635-2644.
doi:10.1021/acscatal.7b04500.
[145] J.R. Mcbride, K.C. Hass, B.D. Poindexter, W.H. Weber, J.R. Mcbride, K.C.
Hass, B.D. Poindexter, W.H. Weber, Raman and X-ray studies of Ce1-xRexO2-y,
where Re=La, Pr, Nd, Eu, Gd, and Tb, J. Appl. Phys. 76 (2006) 2435.
doi:10.1063/1.357593.
[146] A.M.D. Angelo, A.L. Chaffee, Correlations between oxygen uptake and
vacancy concentration in Pr-doped CeO2, ACS Omega, 2 (2017) 2544-2551.
doi:10.1021/acsomega.7b00550.
[147] F. Wang, L. Xu, J. Yang, J. Zhang, L. Zhang, H. Li, Y. Zhao, H. Xing, K.
Wu, G. Qin, W. Chen, Enhanced catalytic performance of Ir catalysts supported
on ceria-based solid solutions for methane dry reforming reaction, Catal. Today.
281 (2017) 295-303. doi:10.1016/j.cattod.2016.03.055.
[148] T. Taniguchi, T. Watanabe, N. Sugiyama, A.K. Subramani, H. Wagata, N.
Matsushita, M. Yoshimura, Identifying defects in ceria-based nanocrystals by UV
resonance Raman spectroscopy, J. Phys. Chem. 113 (2009) 19789-9793. doi:
10.1021/jp9049457.
[149] J.J. Kim, S.R. Bishop, N.J. Thompson, D. Chen, H.L. Tuller, Investigation
of nonstoichiometry in oxide thin films by simultaneous in situ optical absorption
and chemical capacitance measurements: Pr-doped ceria, a case study, Chem.
Doctoral thesis / Yanyan Liu
87
Mater. 26 (2014) 1374-1379. doi:10.1021/cm403066p.
[150] J.J. Kim, S.R. Bishop, D. Chen, H.L. Tuller, Defect chemistry of Pr doped
ceria thin films investigated by in situ optical and impedance measurements, Chem.
Mater. 29 (2017) 1999-2007. doi:10.1021/acs.chemmater.6b03307.
[151] Y.L. Yang, A.J. Jacobson, C.L. Chen, G.P. Luo, K.D. Ross, C.W. Chu,
Oxygen exchange kinetics on a highly oriented La0.5Sr0.5CoO3-δ thin film prepared
by pulsed-laser deposition, Appl. Phys. Lett. 79 (2001) 776-778.
doi:10.1063/1.1390316.
[152] S.B. Adler, Factors governing oxygen reduction in solid oxide fuel cell
cathodes, Chem. Rev. 104 (2004) 4791-4843. doi:10.1021/cr020724o.
[153] H.G. Seo, Y. Choi, W. Jung, Exceptionally enhanced electrode activity of
(Pr,Ce)O2-δ-based cathodes for thin-film solid oxide fuel cells, Adv. Energy Mater.
8 (2018) 1703647. doi:10.1002/aenm.201703647.
[154] R.A. Souza, J.A. Kilner, Oxygen transport in La1-xSrxMn1-yCoyO3±δ
perovskites: Part II. oxygen surface exchange. Solid State Ionics 126 (1999) 153-
161. doi: 10.1016/S0167-2738(99)00228-3.
[155] S.M. Yang, S. Lee, J. Jian, W. Zhang, P. Lu, Q. Jia, H. Wang, T. Won Noh,
S. V. Kalinin, J.L. MacManus‐Driscoll, Strongly enhanced oxygen ion transport
through samarium-doped CeO2 nanopillars in nanocomposite films, Nat. Commun.
6 (2015) 8588. doi:10.1038/ncomms9588.
[156] Y. Wu, C. Xia, W. Zhang, X. Yang, Z.Y. Bao, J.J. Li, B. Zhu, Natural
hematite for next-generation solid oxide fuel cells, Adv. Funct. Mater. (2015).
doi:10.1002/adfm.201503756.
[157] Z. Jiang, Z. Lei, B. Ding, C. Xia, F. Zhao, F. Chen, Electrochemical
characteristics of solid oxide fuel cell cathodes prepared by infiltrating
(La,Sr)MnO3 nanoparticles into yttria-stabilized bismuth oxide backbones, Int. J.
Hydrogen Energy. 35 (2010) 8322-8330. doi:10.1016/j.ijhydene.2009.12.008.
[158] C. Xia, Development of natural mineral composites for low-temperature solid
oxide fuel cells, KTH (2018).
[159] M. Afzal, Semiconductor-ionic materials for low temperature solid oxide fuel
cells, KTH (2019).