close

Вход

Забыли?

вход по аккаунту

?

978-981-10-6617-7

код для вставкиСкачать
Springer Theses
Recognizing Outstanding Ph.D. Research
Yucun Zhou
Study on
Fabrication and
Performance of
Metal-Supported
Solid Oxide Fuel
Cells
Springer Theses
Recognizing Outstanding Ph.D. Research
Aims and Scope
The series “Springer Theses” brings together a selection of the very best Ph.D.
theses from around the world and across the physical sciences. Nominated and
endorsed by two recognized specialists, each published volume has been selected
for its scientific excellence and the high impact of its contents for the pertinent field
of research. For greater accessibility to non-specialists, the published versions
include an extended introduction, as well as a foreword by the student’s supervisor
explaining the special relevance of the work for the field. As a whole, the series will
provide a valuable resource both for newcomers to the research fields described,
and for other scientists seeking detailed background information on special
questions. Finally, it provides an accredited documentation of the valuable
contributions made by today’s younger generation of scientists.
Theses are accepted into the series by invited nomination only
and must fulfill all of the following criteria
• They must be written in good English.
• The topic should fall within the confines of Chemistry, Physics, Earth Sciences,
Engineering and related interdisciplinary fields such as Materials, Nanoscience,
Chemical Engineering, Complex Systems and Biophysics.
• The work reported in the thesis must represent a significant scientific advance.
• If the thesis includes previously published material, permission to reproduce this
must be gained from the respective copyright holder.
• They must have been examined and passed during the 12 months prior to
nomination.
• Each thesis should include a foreword by the supervisor outlining the significance of its content.
• The theses should have a clearly defined structure including an introduction
accessible to scientists not expert in that particular field.
More information about this series at http://www.springer.com/series/8790
Yucun Zhou
Study on Fabrication
and Performance
of Metal-Supported Solid
Oxide Fuel Cells
Doctoral Thesis accepted by
Chinese Academy of Sciences, Shanghai, P.R. China
123
Author
Dr. Yucun Zhou
Shanghai Institute of Ceramics
Chinese Academy of Sciences
Shanghai
P.R. China
Supervisor
Prof. Shaorong Wang
Shanghai Institute of Ceramics
Chinese Academy of Sciences
Shanghai
P.R. China
ISSN 2190-5053
ISSN 2190-5061 (electronic)
Springer Theses
ISBN 978-981-10-6616-0
ISBN 978-981-10-6617-7 (eBook)
https://doi.org/10.1007/978-981-10-6617-7
Library of Congress Control Number: 2017957203
© Springer Nature Singapore Pte Ltd. 2018
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained herein or
for any errors or omissions that may have been made. The publisher remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
Printed on acid-free paper
This Springer imprint is published by Springer Nature
The registered company is Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore
Supervisor’s Foreword
It is my great pleasure to introduce and recommend Dr. Yucun Zhou’s research
work for publication in the series of Springer Theses. His research work focuses on
the design, preparation, and measurements of metal-supported solid oxide fuel cells
(MS-SOFCs), under my supervision.
My research group devoted efforts in developing SOFC technology for distributed supplying of combined power and heat (CHP) applications. The operating
temperature is a key point for this technology, which is deeply related to the cost
and life of SOFC stacks. MS-SOFC is a frontier object in this field, for the cheaper
and more robust properties of porous metals in comparison with the traditional
Ni/YSZ (Yttria-stabilized zirconia) anode. However, the densification of the ceramic electrolyte film on the porous metal by a cheap technology is very difficult.
After joining our group, Dr. Yucun Zhou got to know this issue and began to work
following the tape casting and co-sintering route. We started from cheap SUS430
powder and commercial YSZ powder, trying to get a dense YSZ film supported by
a porous metal layer via the traditional sintering technology. We expected the
shrinkage of the porous metal substrate would help the densification of the electrolyte film. However, the recipe of the tape casting slurry, the sintering process and
controlling of atmospheres, and also the design and optimizing of the microstructure of the electrode active layers should be carefully selected. There are huge
amount of experimental work to be done. Dr. Yucun Zhou has accepted good
training in materials science before, but the determining factor of success is his
smart and level head, as well as his diligence. He stands on the shoulder of others,
keeping modest, reading a lot of literatures to learn ideas for controlling of electrode
structure, and finally progressed much along this way.
v
vi
Supervisor’s Foreword
Part of Yucun Zhou’s research has been published in high-impact journals such
as Advance Energy Materials, Journal of Power Sources, and Journal of The
Electrochemical Society, etc. The publication of this thesis in Springer is believed
to promote scientific research in the community of materials science and technology
in relation to SOFC.
Shanghai, P.R. China
July 2017
Prof. Shaorong Wang
Parts of this thesis have been published in the following journal articles:
(1) Yucun Zhou, Xianshuang Xin, Junliang Li, Xiaofeng Ye, Changrong Xia,
Shaorong Wang, Zhongliang Zhan, Performance and degradation of
metal-supported solid oxide fuel cells with impregnated electrodes, Int.
J. Hydrogen Energy, 2014, 39, 2279–2285.
(2) Yucun Zhou, Chun Yuan, Ting Chen, Xie Meng, Xiaofeng Ye, Junliang Li,
Shaorong Wang, Zhongliang Zhan, Evaluation of Ni and Ni–Ce0.8Sm0.2O2−d
(SDC) impregnated 430L anodes for metal-supported solid oxide fuel cells,
J. Power Sources, 2014, 267, 117–122.
(3) Yucun Zhou, Xiaofeng Ye, Junliang Li, Zhongliang Zhan, Shaorong Wang,
Metal-supported solid oxide fuel cells with a simple structure, J. Electrochem.
Soc., 2014, 161, F332–336.
(4) Yucun Zhou, Ting Luo, Xianlong Du, Jianqiang Wang, Wei Yang, Chunwen
Sun, Changrong Xia, Shaorong Wang, Zhongliang Zhan, High activity of
nanoporous-Sm0.2Ce0.8O2-d@430L composites for hydrogen electro-oxidation
in solid oxide fuel cells, Adv. Energy Mater., 2014, 4.
(5) Yucun Zhou, Ting Chen, Junliang Li, Chun Yuan, Xianshuang Xin, Guoyi
Chen, Guoshuan Miao, Weiting Zhan, Zhan Zhongliang, Wang Shaorong,
Long–term stability of metal–supported solid oxide fuel cells employing
infiltrated electrodes, J. Power Sources, 2015, 295, 67–73.
(6) Weiting Zhan, Yucun Zhou, Ting Chen, Guoshuan Miao, Xiaofeng Ye,
Junliang Li, Zhan Zhongliang, Wang Shaorong, Zhenyan Deng, Long–term
stability of infiltrated La0.8Sr0.2CoO3-d, La0.58Sr0.4Co0.2Fe0.8O3-d and
SmBa0.5Sr0.5Co2.0O5+d cathodes for low temperature solid oxide fuel cells, Int.
J. Hydrogen Energy, 2015, 40, 16532–16539.
(7) Yucun Zhou, Da Han, Chun Yuan, Minquan Liu, Ting Chen, Shaorong Wang,
Zhongliang Zhan, Infiltrated SmBa0.5Sr0.5Co2O5+d cathodes for metal–supported solid oxide fuel cells, Electrochim. Acta, 2014, 149, 231–236.
(8) Yucun Zhou, Hao Wu, Ting Luo, Jianqiang Wang, Yixiang Shi, Changrong
Xia, Shaorong Wang, Zhongliang Zhan, A Nanostructured Architecture for
Reduced-Temperature Solid Oxide Fuel Cells, Adv. Energy Mater., 2015, 5.
vii
Acknowledgements
I would like to thank my teachers, classmates, friends, and families for their
guidance, support, and help over the past five years.
Foremost, I would like to convey my sincere gratitude to my supervisor, Prof.
Shaorong Wang for his meticulous guidance and help. Professor Wang is not only a
wise, diligent, and insightful scholar and engineer, but also a modest and low-key
gentleman. Professor Wang invited me to the area of solid oxide fuel cells, provided
me with tremendous inspiration, encouragement, and support, and taught me to be a
qualified researcher. More importantly, he is a great mentor for my life and taught
me to be a real man. I admired him for his profound knowledge and noble personality. Nothing is enough to show my great appreciation to him. What I only can
do is following his footsteps, to be an aspiring researcher and a useful man.
In addition, I would like to thank Prof. Tinglian Wen for his kind help in my life
and research. I would like to thank Prof. Zhongliang Zhan for his guidance.
I am grateful to many people who have provided assistance in my research work
and brought me friendship and happiness in my daily life. They are Dr. Xiaofeng
Ye, Dr. Junliang Li, Dr. Le Shao, Dr. Da Han, Dr. Juan Zhou, Dr. Ting Luo,
Dr. Yadi Liu, Dr. Chun Yuan, Xie Meng, Dr. Xuejiao Liu, Dr. Jie Zou, Wenzhi
Pan, Dr. Tianyu Zhou, Dr. Yijie Zhou, Dr. Shan Yun, Dr. Haibo Wu, Zhengyi
Zhou, Zhencheng Zhang, Qiang Zhou, Minquan Liu, Guoyi Chen, Xixiang Li,
Guoshuan Miao, Xiaofeng Tong, Ting Chen, Weiting Zhan, and Xiaona Ji.
I would like to thank many researchers and staff in Shanghai Institute of
Ceramics, Chinese Academy of Sciences, for their kind support and help. They are
Dr. Huaiwen Nie, Dr. Xianshuang Xin, Dr. Jian Shi, Fanrong Zeng, Jiqin Qian,
Chongying Zhong, Youpeng Chen, Leimin Liu, Hao Wu, Jun Lu, Yuxin Han, Yide
Sheng, Chucheng Lin, Zhiwei Zhou, Caifei Lu, Xueying Zhao, and Xinhong Lu.
I wish them every success in the future.
I would like to give the highest appreciation to my parents. They raised me by
their hard work. I have learnt many life skills and inherited many excellent moral
ix
x
Acknowledgements
characters from them. Meanwhile, I would like to thank my sisters, brothers-in-law,
and all my relatives for their constant support.
At last, I would like to give a special thanks to my wife Fang Wang. I cannot
thank you enough for your love, support, and standing by me through it all. You are
always in my heart.
Contents
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
1
1
2
3
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
3
6
7
8
10
11
12
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
15
15
16
16
17
18
18
19
19
......
24
1 Research Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.1 Solid Oxide Fuel Cell . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.1.1 Introduction of Solid Oxide Fuel Cell . . . . . . . . . . . .
1.1.2 Operating Principle of Solid Oxide Fuel Cell . . . . . .
1.2 Metal-Supported Solid Oxide Fuel Cell . . . . . . . . . . . . . . . .
1.2.1 Introduction of Metal-Supported Solid Oxide
Fuel Cell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.2 Choice of the Metal Support . . . . . . . . . . . . . . . . . .
1.2.3 Materials and Fabrication Techniques of Electrolytes .
1.2.4 Anode Issues and Corresponding Strategies . . . . . . . .
1.2.5 Cathode Issues and Corresponding Strategies . . . . . .
1.3 Scope of This Thesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2 Fabrication and Investigation of Intermediate-Temperature
MS–SOFCs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2 Experimental Section . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2.1 Technical Route . . . . . . . . . . . . . . . . . . . . . . . . .
2.2.2 Fabrication of Symmetric and Single Cells . . . . . .
2.2.3 Material Characterizations . . . . . . . . . . . . . . . . . .
2.2.4 Electrochemical Measurements . . . . . . . . . . . . . . .
2.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.1 Investigation of Infiltrated LSFSc–YSZ Cathodes .
2.3.2 Investigation of Infiltrated Ni–430L Anodes
and the MS–SOFCs . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
xi
xii
Contents
2.3.3 Investigation of Infiltrated SDC–430L Anodes
and the MS–SOFCs . . . . . . . . . . . . . . . . . . . . . . .
2.3.4 Investigation of Infiltrated Ni–SDC–430L Anodes
and the MS–SOFCs . . . . . . . . . . . . . . . . . . . . . . .
2.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
......
32
......
......
......
43
55
56
.
.
.
.
.
.
.
.
.
.
.
.
.
.
61
61
62
62
62
62
63
......
63
......
......
......
70
75
75
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
77
77
78
78
78
79
79
79
.......
.......
.......
83
88
89
5 Summary and Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2 Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
91
91
92
3 Fabrication and Investigation of Low-Temperature
MS–SOFCs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2 Experimental Section . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2.1 Fabrication of Symmetric and Single Cells . . . . . .
3.2.2 Material Characterizations . . . . . . . . . . . . . . . . . .
3.2.3 Electrochemical Measurements . . . . . . . . . . . . . . .
3.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . .
3.3.1 Investigation of Infiltrated LSC/LSCF/SBSC–SSZ
Cathodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.3.2 Investigation of Infiltrated SBSC–SSZ Cathodes
and the MS–SOFCs . . . . . . . . . . . . . . . . . . . . . . .
3.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4 Fabrication and Investigation of MS–SOFCs
with a Symmetric Configuration . . . . . . . . . . . . . . . . . . . . .
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2 Experimental Section . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.1 Fabrication of Symmetric and Single Cells . . . . .
4.2.2 Material Characterizations . . . . . . . . . . . . . . . . .
4.2.3 Electrochemical Measurements . . . . . . . . . . . . . .
4.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . .
4.3.1 Investigation of Infiltrated SBSC–430L Cathodes
4.3.2 Investigation of MS–SOFCs with a Symmetric
Configuration . . . . . . . . . . . . . . . . . . . . . . . . . .
4.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Chapter 1
Research Background
1.1
1.1.1
Solid Oxide Fuel Cell
Introduction of Solid Oxide Fuel Cell
Fuel cells are electrochemical energy conversion devices which directly convert the
chemical energy in the fuels (e.g., hydrogen, methane) into electricity. Unlike other
types of chemical batteries, in which the chemical energy for electricity conversion
is obtained by consuming the electrode materials, the chemical energy in fuel cells
is derived from the reaction energy of fuels and oxidants. Thus, fuel cells can be
regard as “chemical generators”, which can generate electricity uninterruptedly by a
sustained supply of fuels and oxidants. Some characteristics of fuel cells are list here
[1, 2]:
(1) High efficiency: Since no combustion reaction or transmission equipment is
needed in the energy conversion process for the fuel cells, the efficiency is not
limited by the Carnot cycle, and the power generating efficiency can be as high
as 40–65%. If combined with heat, the efficiency can reach ˃90%.
(2) Environmentally friendly: The energy conversion process of the fuel cells is
achieved by the electrochemical process without combustion reaction or
moving component, thus no noise, dust or toxic gases like SOx, NOx generated.
(3) Flexibility of scale and usage: Different from traditional batteries, in which the
power and capacity are often convoluted, fuel cells allow easy independent
scaling between power and capacity, thus scale well from the 1–W range to the
megawatt range. Due to the flexibility of output power and scale, fuel cells have
a wide application from small portable power source to large fixed power
station.
Fuel cells are composed of a porous cathode where the reduction reaction of the
oxidants (e.g., oxygen) occurs, a dense electrolyte providing ionic conducting, and
a porous anode where the oxidation reaction of the fuels happens. In general, the
© Springer Nature Singapore Pte Ltd. 2018
Y. Zhou, Study on Fabrication and Performance of Metal-Supported Solid Oxide
Fuel Cells, Springer Theses, https://doi.org/10.1007/978-981-10-6617-7_1
1
2
1 Research Background
property and the application of the fuel cells are determined by the property of the
electrolytes. Based on different types of electrolytes, fuel cells can be divided into
Alkaline Fuel Cell (AFC), Proton Exchange Membrane Fuel Cell (PEMFC),
Phosphoric Acid Fuel Cell (PAFC), Molten Carbon Fuel cell (MCFC) and Solid
Oxide Fuel Cell (SOFC). The SOFC consists of a ceramic, oxide ionic conducting
electrolyte sandwiched by a ceramic anode and a ceramic cathode. The key features
of the SOFC are its all solid state construction and high operating temperature
(600–1000 °C). The combination of these features leads to a number of distinctive
and attractive attributes for the SOFC including noble metals free, no leakage or
corrosion risk of the liquid electrolyte, cell and stack design flexibility, multi-fuel
capability (including hydrocarbon fuels), and high efficiency [3–6]. The SOFC has
been considered for a broad spectrum of power generation applications, including
power systems ranging from watt range to megawatt range, e.g., Auxiliary Power
Unit (APU) for vehicles, Combined Heat and Power (CHP) system for buildings,
and large-scale power plants. Due to these merits, the SOFC has became one of the
most consistent developed fuel cells (1950s to nowadays), and has demonstrated its
commercial potential in the past several years.
1.1.2
Operating Principle of Solid Oxide Fuel Cell
The operation principle of a SOFC is schematically shown in Fig. 1.1. During the
operating process of a SOFC, oxygen is adsorbed on the surface of the porous
cathode and reduced into oxide ions under the presence of electrons from an
external circuit. These oxide ions migrate through the dense solid electrolyte to the
anode side driven by the oxygen chemical potential gradient. In the porous anode,
Fig. 1.1 A schematic
diagram of the SOFC
1.1 Solid Oxide Fuel Cell
3
oxide ions react with fuels, H2 or CO, to produce H2O or CO2, and the released
electrons flow to the external circuit.
When using hydrogen as the fuel, reaction in the anode is:
2O2 þ 2H2 ! 2H2 O þ 4e
ð1:1Þ
Reaction in the cathode is:
O2 þ 4e ! 2O2
ð1:2Þ
Full reaction of the fuel cell is:
2H2 þ O2 ! 2H2 O
ð1:3Þ
The open-circuit voltage, EOCV, of the cell can be calculated from the free energy
change, DG, of the electrochemical reaction or from the partial pressure of the
oxygen PO(c) at the cathode and PO(a) at the anode:
EOCV ¼ DG=nF ¼ ðRT=nFÞ ln PO ðcÞ=PO ðaÞ
ð1:4Þ
Here, R is the gas constant, T is the absolute temperature, F is the Faraday
constant, and n is the electron equivalent of oxygen (n = 4).
1.2
1.2.1
Metal-Supported Solid Oxide Fuel Cell
Introduction of Metal-Supported Solid Oxide
Fuel Cell
As introduced above, a SOFC is composed of three components: the anode, electrolyte and cathode. A typical electrolyte material of a SOFC is yttria-stabilized
zirconia (YSZ), an oxide ion conductor at elevated temperatures. The anode is
usually a nickel-zirconia cermet (Ni–YSZ), and the cathode a perovskite material,
e.g., strontium doped lanthanum manganite (LSM). In practical application, one of
the three components should be thick enough to provide mechanical support for the
whole cell. Based on various supports, the SOFC can be divided into electrolytesupported SOFC (ES–SOFC), anode-supported SOFC (AS–SOFC), cathodesupported SOFC (CS–SOFC) and metal-supported SOFC (MS–SOFC) (Fig. 1.2).
Properties of different SOFC configurations are listed in Table 1.1.
The early development of SOFC mainly focused on the ES–SOFC. Due to the
large ohmic impedance of the thick electrolyte (100–1000 µm), operating temperatures of the ES–SOFC should be as high as 850–1000 °C. Such high temperatures restrict the selection of materials used in the stack and the balance-of-plant
(BOP), bring challenges like thermal insulation and high-temperature oxidation/
corrosion. In another aspect, the ES–SOFC still exhibits a number of advantages
4
1 Research Background
Fig. 1.2 A schematic diagram of different SOFC configurations
Table 1.1 A comparison of different SOFC configurations
SOFC
configurations
Advantages
Disadvantages
ES–SOFC
Good redox stability and long-term
durability, Facile fabrication
AS–SOFC
Low operating temperature
High output power
High redox stability of the anode
Low cost, High mechanical strength,
Excellent redox and thermal shock
tolerance
Large ohmic resistance
High operating temperature
High cost
Low redox stability
CS–SOFC
MS–SOFC
Low output power
Densification issue of the
electrolyte, Oxidation of the metal,
Cr poisoning issue
including: high stability of the structure and performance, good redox tolerance and
facile fabrication.
For the AS–SOFC, a thick anode (300–1000 µm) is used to support the whole
cell and the thickness of the electrolyte can be reduced to less than 50 µm. By using
a thin electrolyte, the ohmic resistance of the fuel cell is reduced significantly,
allowing the operation of an AS–SOFC at the intermediate temperatures (600–
800 °C). In addition, reduced operating temperatures allow for the application of a
wider range of materials and more cost-effective fabrication process, particularly in
relation to the interconnectors and BOP [7].
Compared with the AS–SOFC, the CS–SOFC shows a much better redox tolerance due to the thinner anode which can avoid the significant volume change of
1.2 Metal-Supported Solid Oxide Fuel Cell
5
the anode resulting from redox cycles. Siemens/Westinghouse has successfully
developed the tubular CS–SOFC technology and demonstrated the world’s first
highly efficient, longest running (over 13,000 h) 100 kW SOFC-CHP system and
the first highest–efficiency, 220 kW class pressurized SOFC-gas turbine
(PSOFC-GT) hybrid system based on this technology [8, 9].
The above cell configurations (ES–SOFC, AS–SOFC and CS–SOFC) using
porous ceramics or cermets as the mechanical supports bring challenges like high
cost of the raw materials and the fabricating processes, poor thermal and electrical
conductivity, and inferior mechanical strength due to the inherent properties of the
ceramics. In the last decade, benefitting from the development of new materials and
techniques, the operating temperatures of the SOFC have been reduced to lower
than 800 °C, under which the oxidation issue of metals, e.g., stainless steels can be
significantly relieved [10–15]. Thus, the MS–SOFC using a porous metal as the cell
support regained people’s interest recently. The MS–SOFC shows a number of
advantages over the traditional all–ceramic structured SOFC:
1. High mechanical strength: Due to the high mechanical strength especially the
fracture toughness of the metals, MS–SOFC exhibits much higher mechanical
ruggedness than that of the traditional fragile all-ceramic structured SOFC [16].
2. Increased thermal cycling stability: The high ductility and thermal conductivity
of the metals can help to relieve the mechanical stress and thermal stress in the
cells or stacks, thus to increase the thermal shock resistance of the SOFC. As
reported, the ability of a MS–SOFC to withstand rapid thermal cycling between
200 and 800 °C at 50 °C min−1 has been demonstrated [17].
3. Excellent redox tolerance: For the most widely developed AS–SOFC, a volumetric expansion of more than 40% is observed for the Ni particles when
oxidized to NiO. The volumetric expansion will cause a expansion strain in the
anode, resulting in a tensile strain in the electrolyte and a respective stress,
leading to the fracture of the electrolyte and the failure of fuel cells [18, 19].
While for the MS–SOFC, stainless steels with good redox stability are used as
the thick supporting layer and the thickness of the Ni based anode has been
greatly reduced, thus an excellent redox tolerance is guaranteed. For example,
the MS–SOFC developed in Ceres Power can withstand more than 100 accelerated redox cycles without degradation in performance [20].
4. Low cost: In the MS–SOFC, the cost of the raw materials can be greatly reduced
by replacing the expensive rare earth oxides with much cheaper metals, e.g.,
ferritic stainless steel. In addition, facile and low-cost manufacturing techniques,
e.g., welding, can be expected to be applied in the fabrication of MS–SOFCs,
which further reducing the manufacturing cost [21, 22].
In all, due to the high mechanical strength, good ductility, excellent thermal and
electrical conductivity, and good redox stability of the alloys, using metals, e.g.,
alloys as the cell substrate can effectively circulate the inherent drawbacks of the
ceramic substrate and the technical challenges of SOFCs. Based on these merits,
MS–SOFC has been regard as one of the most promising cell configurations for
6
1 Research Background
power generation systems [23]. However, before the successful commercial of the
MS–SOFC, there are still a number of challenges to be solved, e.g., the densification issue of the ceramic electrolyte, oxidation of the metal support, elements
inter-diffusion between the FeCr substrate and the Ni-containing anode, and Cr
poisoning issue of the cathode.
1.2.2
Choice of the Metal Support
The selection of proper metal supports is critical for the successful fabrication and
stable operation of a MS–SOFC. In general, metal supports for a MS–SOFC should
fulfill the following requirements: (1) high mechanical strength, (2) high electrical
conductivity, (3) good high–temperature oxidation and corrosion resistance under
the SOFC operating environment (600–800 °C, oxidizing and reducing atmospheres), (4) matched thermal expansion coefficients (TECs) with other cell components, (5) good manufacturability, (6) low cost.
Metals, especially chromia-forming alloys have been widely investigated as
interconnect materials for SOFCs. Due to the similar operating condition and
requirement, in principle, the selection of proper metal supports can refer to that of
metal interconnects. The most widely used alloys for the interconnects are composed of Cr–Fe–Ni three phases [24]. Specifically [25, 26]: chromium based alloys
have an excellent high-temperature oxidation resistance (900–1000 °C) due to the
high-conductive protective oxides generated on the surface. However, the high
chromium content in the Cr based alloys will cause the issues like chromium
poisoning of cathode and excessive chromia growth. In addition, the Cr based
alloys are difficult and costly to fabricate. By reducing the Cr content, the Fe–
Cr-based alloys exhibit an enhanced manufacturability and a reduced cost. To
ensure the formation of a continuous, protective Cr2O3 scale, the critical minimum
Cr content is approximately 20–25%. Based on compositions, stainless steels are
usually divided into four groups: (i) ferritic steels, (ii) austenitic steels,
(iii) martensitic steels and (iv) precipitation hardening steels [24]. Among them, the
ferritic stainless steels, e.g., SS430L, SS440L and SS410L with a Cr content of 15–
30% are the most promising candidates for MS–SOFC applications due to the
matched TEC with those of other SOFC materials caused by their body-centered
cubic structure [14, 27–29]. Ferritic stainless steels also have good oxidation and
corrosion resistance, low cost and good manufacturability. The drawback of such
Fe–Cr-based alloys lies in their low mechanical strength. Austenitic stainless steels
with a Cr content higher than 30%, a Ni content of 8%, and trace Mo, Ti and N
have high mechanical strength, good oxidation resistance, good manufacturability,
and low cost. While the high TEC of such alloys affects their wide application in
MS–SOFCs. Compared with Fe–Cr-based alloys, Ni–Cr-base alloys always
demonstrate higher mechanical strength, better oxidation resistance and scale
electrical conductivity. However, as the austenitic stainless steels, a major drawback
1.2 Metal-Supported Solid Oxide Fuel Cell
7
of these alloys is their high TEC. In addition, Ni and Ni–Fe alloys which can be
easily fabricated by the well established manufacturing technology for SOFCs have
also been investigated as proper supporting materials for MS–SOFCs [30–32].
However, these materials have many disadvantages like poor redox resistance, high
cost, low oxidation resistance, and relatively low mechanical strength.
To date, a number of metals have been applied in MS–SOFCs as potential
supporting materials. It’s hard to say which will be the most promising one since
other factors like the configuration of the cell, materials of other components and
manufacturing techniques should also be taken into consideration.
1.2.3
Materials and Fabrication Techniques of Electrolytes
Well-established and low-cost fabrication techniques for ceramic membranes like
tape casting, screen printing and suspend fluid spray have been widely applied in the
electrolyte fabrication for traditional all-ceramic SOFCs. However, high sintering
temperatures (1200–1500 °C) are needed to densify the electrolytes by these
wet-chemical methods, and severe oxidation of the metal supports will occur under
such high temperatures. Co-firing the metal support and electrolyte layer in a
reducing atmosphere is a good way to protect the metal, while electrolytes like doped
ceria and strontium and magnesium doped lanthanum gallate (LSGM) are not stable
under the reducing atmosphere at high temperatures. Figure 1.3 shows the X–Ray
diffraction (XRD) patterns of the LSGM powder after calcining in 95% N2–5% H2 at
1000 and 1200 °C, respectively. It can be found that no obvious change in the
perovskite phase is observed after treating at 1000 °C, while a full decomposition
into La2O3 and SrLaGaO4 occurs at a higher temperature of 1200 °C. Zirconia based
electrolytes, e.g., YSZ, has a good ionic conductivity, chemical and structure stability under a wide range of partial pressure of oxygen. This kind of electrolyte can
Fig. 1.3 XRD patterns of the LSGM powder after calcining in 95% N2–5% H2 at: a 1000 °C and
b 1200 °C
8
1 Research Background
be fabricated by the wet-chemical forming and high-temperature, reducing–atmosphere sintering method [33, 34]. As reported, a MS–SOFC has been developed by
laminating the tape casted Fe22Cr stainless steel layer, the Fe22Cr+0–50 vol.%YSZ
cermet layer and the Sc, Y co-doped ZrO2 (ScYSZ) layer, followed by a
high-temperature sintering process under the reducing atmosphere (H2/Ar) [35].
Low-temperature techniques, e.g., plasma spray, pulsed laser deposition
(PLD) and electrostatic and pneumatic spray deposition have also been applied to
fabricate electrolyte membranes for MS–SOFCs. A major advantage of these techniques is the low operating temperature which can effectively avoid the oxidation of
metal supports. Researchers from the National Research Council Canada fabricated a
dense samarium doped ceria (SDC)/scandia stabilized zirconia (ScSZ) bi-layer
electrolyte onto the porous SS430L substrate by a combination of PLD and wet
ceramic processes [28]. Researchers from the same group have also obtained dense
SDC electrolyte layers on the SS430L and Hastelloy X substrates by the spray
pyrolysis and thermal spray method, respectively [36, 37]. Hwang et al. used the
atmospheric plasma spray (APS) technology to deposit a La0.8Sr0.2Ga0.8Mg0.2O3
(LSGM) electrolyte layer onto the porous nickel substrate and the resulting MS–
SOFC showed an output power density of 1.27 W cm−2 at 800 °C [38]. Ju et al.
fabricated a SDC/LAGM bi-layer on the Ni–Fe metal support by the PLD method
and the resulting MS–SOFC exhibited high power densities of 1.99, 1.1 and
0.53 W cm−2 at 700, 600 and 500 °C, respectively [39].
Although some promising results have been demonstrated, the above
physical/chemical deposition techniques have several drawbacks: for the PLD
method, a high vacuum is needed, uneven deposition for large samples is reported
together with a high cost; for the spray pyrolysis, a strict requirement of the porous
substrate (a low surface roughness and a small pore size) is needed and the density
of the fabricated electrolytes is not high enough; for the plasma spray method, low
pO2 during deposition results in strain relaxation in post annealing and formation of
micro-cracks, the electrolyte thickness is typically higher than 30 lm to be gas tight
[40]. Thus, it is urgent to develop a low-cost, high efficient and facile technique to
fabricate large sized electrolytes with a high conductivity for MS–SOFCs.
1.2.4
Anode Issues and Corresponding Strategies
Traditional MS–SOFCs have a four-layer structure with a cell configuration of
“metal support-anode-electrolyte-cathode” and the typical anode materials are Ni
based cermets. Villarreal et al. fabricated a MS–SOFC with a Fe–Cr metal support,
a Ni–YSZ anode and a thin YSZ electrolyte by sintering the multi-layers in a
reducing atmosphere at 1350 °C [33]. However, the aggregation and coarsening of
the Ni particles caused by the high-temperature sintering process leaded to an
excessive densification of the Ni–YSZ anode, which resulted in a low power
density of the MS–SOFCs (0.1 W cm−2 at 800 °C) [33]. In addition, when the Ni
based anode and the ferritic FeCr substrate contact directly, a mutual diffusion
1.2 Metal-Supported Solid Oxide Fuel Cell
9
process of Fe, Ni, and Cr on the substrate/anode interface will occur during cell
fabrication as well as during electrochemical operation of the MS–SOFCs. The
diffusion process will bring two issues:
1. The diffusion of Ni from the anode to the FeCr substrate will alter the substrate
structure from the ferritic to the austenitic structure, leading to an increased TEC
of the substrate. The increased TEC can cause internal cracks, even structural
failure of the whole fuel cell. Furthermore, the oxidation resistance of the
substrate matrix will be changed by forming a new alloy and the long-term
stability will be influenced [41].
2. The diffusion of Fe and Cr from the substrate to the anode will convert the Ni
phase into a Ni-based alloy with relatively high Cr and Fe contents. This conversion will inhibit the electrochemical activity of the Ni based anode, resulting
in an increased polarisation resistance and hence a decreased performance as
well as long term stability [41].
Brandner et al. investigated the interfacial diffusion behavior of elements
between the Crofer 22 APU alloy and the Ni layer, and the diffusion depth of Cr can
be as long as 70 lm into the Ni layer after the treatment at 1100 °C for 3 h [42].
Franco et al. reported that a diffusion depth of 15–20 lm for element Ni, Fe and Cr
was found after the 200 h operation of MS–SOFCs at 800 °C [41].
Several strategies have been explored to solve the Ni coarsening and elemental
mutual diffusion issues:
1. Low fabricating and operating temperatures. Using low-temperature
physical/chemical deposition techniques to fabricate MS–SOFCs can effectively avoid the Ni coarsening and elemental diffusion issues [43, 44]. As
reported by the Ceres Power, dense ceria based electrolytes can be achieved at
only 1000 °C, at which the oxidation of metal supports and coarsening of Ni
based anodes can be greatly inhibited [45]. Furthermore, due to the low operating temperature of 600 °C, elemental diffusion behavior is not obvious, thus a
long-term stability of the performance can be guaranteed.
2. Alternative anode designs. Placing the metal support in the cathode side is
another way to solve the elemental diffusion issue. Waldbillig et al. deposited a
LSM–YSZ cathode layer onto the porous SS430L support, followed by the YSZ
electrolyte layer and the NiO–YSZ anode layer onto the cathode. This strategy
can avoid the diffusion issue between the metal support and the Ni based anode
[46]. However, Cr poisoning of the cathode materials will be another serious
issue.
3. Diffusion barrier layers. Insertion of a diffusion barrier layer (DBL) like CeO2
and La0.6Sr0.2Ca0.2CrO3 between the metal support layer and the anode layer has
been widely explored as a solution to Ni and Fe/Cr inter-diffusion [41, 42, 47,
48]. As reported, with a DBL, the degradation of a MS–SOFC has been reduced
from ˃20% (without a DBL) to less than 1% when operated at 800 °C and 0.7 V
during the initial 1000 h [41]. For a DBL, besides the diffusion blocking effect
of Fe, Cr and Ni, it should have an appropriate porosity, a good electronic
10
1 Research Background
conductivity, a similar TEC to other components and a good chemical compatibility with other materials.
4. Infiltrated anode design. The infiltration/impregnation technique which is conducted by infiltrating the precursor solution of the catalyst into a pre-sintered
porous backbone, followed by a low-temperature heat treatment to convert the
precursor solution into nanoparticles has been widely used in the fabrication of
SOFCs [49]. In contrast to that of the conversional sintering method, the fabricating temperature of the infiltration method can be remarkably reduced from
1000–1400 °C to 700–850 °C. Thus, Ni coarsening and elemental mutual diffusion issues during cell fabrication can be effectively avoided. Ni and Ni-doped
CeO2 anodes have been fabricated by the infiltration method and promising
electrochemical performances have been achieved [50, 51].
1.2.5
Cathode Issues and Corresponding Strategies
Traditional cathode materials such as LSM and La1–xSrxFeO3−d (LSF) should be
sintered in air in the temperature range of 1000–1200 °C to get a good adhesion
with the electrolyte and an acceptable electrochemical performance, while the
stainless steel substrate would suffer excessive oxidation at so high temperatures.
A reducing atmosphere can protect the steel substrate while the decomposition of
these cathode materials would occur in such atmosphere. As shown in Fig. 1.4,
after treating at 1000 °C for 10 h, the (La0.8Sr0.2)0.95MnO3 (LSM) powder has been
thoroughly decomposed into MnO, La2O3 and MnLa2O4.
Using physical deposition techniques (e.g., plasma spray) can circumvent this
issue [52]. However, due to the high cost of the deposition techniques, most
researchers choose the in-situ sintering method (sintering the cathode during the cell
testing process) to fabricate cathodes for MS–SOFCs [53–55]. Kim et al. investigated the in-situ sintered Ba0.5Sr0.5Co0.8Fe0.2O3−d (BSCF) cathode and a good
sinterability was demonstrated [56]. SmBa0.5Sr0.5Co2.0O5−d (SBSCO) has also been
Fig. 1.4 XRD patterns of the
LSM powder after calcining
in 95% N2–5% H2 at 1000 °C
1.2 Metal-Supported Solid Oxide Fuel Cell
11
applied in MS–SOFCs as the in-situ sintered cathode and a power density of
0.5 W cm−2 (800 °C) for the single cell was achieved [57]. However, the poor
chemical compatibility between those in-situ sintered cathodes and the zirconia
based electrolytes will reduce the stability of the fuel cells. Similar to the anodes,
the cathode issues can also be solved by using the infiltration method. LSM infiltrated YSZ cathode has been applied in the MS–SOFC and a good performance has
been achieved at 650–750 °C [50].
1.3
Scope of This Thesis
In this thesis, we aim to develop novel MS–SOFCs using low-cost and mechanical
robust stainless steels (430L) replacing the ceramic materials as the supports for
SOFCs. In order to solve the issues during the cell fabrication and operation processes, and enhance the electrochemical performance and stability of MS–SOFCs, a
“tape casting-sintering-infiltrating” method and a “micro-nano” structure were
developed. Besides, the structure-performance relationship of the electrodes, reaction kinetics of the electrodes and degradation mechanisms of the fuel cells were
also investigated.
1. Fabrication and investigation of intermediate-temperature MS–SOFCs. In order
to enhance the performances of intermediate-temperature MS–SOFCs (600–
800 °C), La0.6Sr0.4Fe0.9Sc0.1O3−d (LSFSc) cathode and Ni-based anode materials were applied by the infiltration method to reduce the polarization resistances of cathodes and anodes, respectively. The structure, phase and
morphology of the electrodes have been characterized; the loading, heat-treating
temperature and composition of the infiltrated electrodes have been optimized;
the reaction kinetics of the electrodes and the degradation mechanism of the
MS–SOFCs have been studied.
2. Fabrication and investigation of low-temperature MS–SOFCs. In order to
enhance the performances of low-temperature MS–SOFCs (<600 °C), polarization resistances and long-term stabilities of the infiltrated La0.8Sr0.2CoO3−d
(LSC)-scandia stabilized zirconia (SSZ), La0.58Sr0.4Co0.2Fe0.8O3−d (LSCF)-SSZ
and SmBa0.5Sr0.5Co2.0O5+d (SBSC)-SSZ cathodes were investigated. MS–
SOFCs with SBSC-SSZ cathodes have been fabricated and the electrochemical
performance, long-term stability and thermal shock resistance have been
evaluated.
3. Fabrication and investigation of MS–SOFCs with a symmetric configuration. In
order to simplify the structure and enhance the performance of the MS–SOFCs,
a symmetric configuration with “anode infiltrated 430L-electrolyte-cathode
infiltrated 430L” was developed. Reaction kinetics of the cathode, and the
electrochemical performance and stability of the MS–SOFCs have been
evaluated.
12
1 Research Background
References
1. O’hayre R, Cha S-W, Prinz FB, Colella W (2016) Fuel cell fundamentals. Wiley, New York
2. McIntosh S, Gorte RJ (2004) Direct hydrocarbon solid oxide fuel cells. Chem Rev 104:4845–
4865
3. Minh NQ (2004) Solid oxide fuel cell technology—features and applications. Solid State
Ionics 174:271–277
4. Ormerod RM (2003) Solid oxide fuel cells. Chem Soc Rev 32:17–28
5. Singhal SC (2002) Solid oxide fuel cells for stationary, mobile, and military applications.
Solid State Ionics 152:405–410
6. Minh NQ, Mogensen MB (2013) Reversible solid oxide fuel cell technology for green fuel
and power production. Electrochem Soc Interface 22:55
7. Aguiar P, Adjiman CS, Brandon NP (2004) Anode-supported intermediate temperature direct
internal reforming solid oxide fuel cell. I: model-based steady-state performance. J Power
Sources 138:120–136
8. George RA (2000) Status of tubular SOFC field unit demonstrations. J Power Sources
86:134–139
9. Huang K, Singhal SC (2013) Cathode-supported tubular solid oxide fuel cell technology: a
critical review. J Power Sources 237:84–97
10. Shao ZP, Haile SM (2004) A high-performance cathode for the next generation of solid-oxide
fuel cells. Nature 431:170–173
11. Wachsman ED, Lee KT (2011) Lowering the temperature of solid oxide fuel cells. Science
334:935–939
12. Zhan Z, Bierschenk DM, Cronin JS, Barnett SA (2011) A reduced temperature solid oxide
fuel cell with nanostructured anodes. Energy Environ Sci 4:3951–3954
13. Lee JG, Park JH, Shul YG (2014) Tailoring gadolinium-doped ceria-based solid oxide fuel
cells to achieve 2 W cm−2 at 550 °C. Nat Commun 5:4045
14. Molin S, Kusz B, Gazda M, Jasinski P (2008) Evaluation of porous 430L stainless steel for
SOFC operation at intermediate temperatures. J Power Sources 181:31–37
15. Antepara I, Villarreal I, Rodriguez-Martinez LM, Lecanda N, Castro U, Laresgoiti A (2005)
Evaluation of ferritic steels for use as interconnects and porous metal supports in IT–SOFCs.
J Power Sources 151:103–107
16. Leah RT, Brandon NP, Aguiar P (2005) Modelling of cells, stacks and systems based around
metal-supported planar IT–SOFC cells with CGO electrolytes operating at 500–600 °C.
J Power Sources 145:336–352
17. Matus YB, De Jonghe LC, Jacobson CP, Visco SJ (2005) Metal-supported solid oxide fuel
cell membranes for rapid thermal cycling. Solid State Ionics 176:443–449
18. Malzbender J, Wessel E, Steinbrech RW (2005) Reduction and re-oxidation of anodes for
solid oxide fuel cells. Solid State Ionics 176:2201–2203
19. Young JL, Birss VI (2011) Crack severity in relation to non-homogeneous Ni oxidation in
anode-supported solid oxide fuel cells. J Power Sources 196:7126–7135
20. Leah R, Lankin M, Bone A, Selcuk A, Pierce R, Rees L et al (2013) Towards a fully REDOX
stable SOFC: cell development at Ceres Power. ECS Trans 57:849–856
21. Tucker MC, Jacobson CP, De Jonghe LC, Visco SJ (2006) A braze system for sealing
metal-supported solid oxide fuel cells. J Power Sources 160:1049–1057
22. Sakuno S, Takahashi S, Sasatsu H (2009) Metal-supported SOFC development at JPOWER.
ECS Trans 25:731–737
23. Tucker MC (2010) Progress in metal-supported solid oxide fuel cells: a review. J Power
Sources 195:4570–4582
24. Yang Z, Weil KS, Paxton DM, Stevenson JW (2003) Selection and evaluation of
heat-resistant alloys for SOFC interconnect applications. J Electrochem Soc 150:A1188–
A1201
References
13
25. Wu J, Liu X (2010) Recent development of SOFC metallic interconnect. J Mater Sci Technol
26:293–305
26. Fergus JW (2005) Metallic interconnects for solid oxide fuel cells. Mater Sci Eng, A
397:271–283
27. Belogolovsky I, Hou PY, Jacobson CP, Visco SJ (2008) Chromia scale adhesion on 430
stainless steel: effect of different surface treatments. J Power Sources 182:259–264
28. Hui S, Yang D, Wang Z, Yick S, Deces-Petit C, Qu W et al (2007) Metal-supported solid
oxide fuel cell operated at 400–600 °C. J Power Sources 167:336–339
29. Yoo Y, Wang Y, Deng X, Singh D, Legoux J-G (2012) Metal supported tubular solid oxide
fuel cells fabricated by suspension plasma spray and suspension high velocity oxy-fuel spray.
J Power Sources 215:307–311
30. Kong Y, Hua B, Pu J, Chi B, Jian L (2010) A cost-effective process for fabrication of
metal-supported solid oxide fuel cells. Int J Hydrogen Energy 35:4592–4596
31. Li K, Wang X, Jia L, Yan D, Pu J, Chi B et al (2014) High performance Ni–Fe alloy
supported SOFCs fabricated by low cost tape casting-screen printing-cofiring process. Int J
Hydrogen Energy 39:19747–19752
32. Ju Y-W, Ida S, Inagaki T, Ishihara T (2011) Reoxidation behavior of Ni–Fe bimetallic anode
substrate in solid oxide fuel cells using a thin LaGaO3 based film electrolyte. J Power Sources
196:6062–6069
33. Villarreal I, Jacobson C, Leming A, Matus Y, Visco S, De Jonghe L (2003) Metal-supported
solid oxide fuel cells. Electrochem Solid State Lett 6:A178–A179
34. Villarreal I, Rivas M, Rodriguez-Martinez LM, Otaegi L, Zabala A, Gomez N et al (2009)
Tubular metal supported SOFC development for domestic power generation. ECS Trans
25:689–694
35. Blennow P, Hjelm J, Klemenso T, Persson AH, Ramousse S, Mogensen M (2011) Planar
metal-supported SOFC with novel cermet anode. Fuel Cells 11:661–668
36. Xie Y, Neagu R, Hsu C-S, Zhang X, Deces-Petit C, Qu W et al (2010) Thin film solid oxide
fuel cells deposited by spray pyrolysis. J Fuel Cell Sci Technol 7:021007
37. Hui R, Berghaus JO, Deces-Petit C, Qu W, Yick S, Legoux J-G et al (2009) High
performance metal-supported solid oxide fuel cells fabricated by thermal spray. J Power
Sources 191:371–376
38. C-s Hwang, Tsai C-H, Yu J-F, Chang C-L, Lin J-M, Shiu Y-H et al (2011) High performance
metal-supported intermediate temperature solid oxide fuel cells fabricated by atmospheric
plasma spraying. J Power Sources 196:1932–1939
39. Ju Y-W, Inagaki T, Ida S, Ishihara T (2011) Sm(Sr)CoO3 cone cathode on LaGaO3 thin film
electrolyte for IT–SOFC with high power density. J Electrochem Soc 158:B825–B830
40. Larring Y, Fontaine M-L (2013) Critical issues of metal-supported fuel cell. In: Solid oxide
fuels cells: facts and figures. Springer, Berlin, pp 71–93
41. Franco T, Schibinger K, Ilhan Z, Schiller G, Venskutonis A (2007) Ceramic diffusion barrier
layers for metal supported SOFCs. In: Eguchi K, Singhai SC, Yokokawa H, Mizusaki H
(eds) Solid oxide fuel cells, vol 10, pp 771–80
42. Brandner M, Bram M, Froitzheim J, Buchkremer HP, Stoever D (2008) Electrically
conductive diffusion barrier layers for metal-supported SOFC. Solid State Ionics 179:1501–
1504
43. Schiller G, Henne R, Lang M, Müller M (2004) Development of solid oxide fuel cells by
applying DC and RF plasma deposition technologies. Fuel Cells 4:56–61
44. Ishihara T, Yan J, Shinagawa M, Matsumoto H (2006) Ni–Fe bimetallic anode as an active
anode for intermediate temperature SOFC using LaGaO3 based electrolyte film. Electrochim
Acta 52:1645–1650
45. Bance P, Brandon NP, Girvan B, Holbeche P, O’Dea S, Steele BCH (2004) Spinning-out a
fuel cell company from a UK University—2 years of progress at Ceres Power. J Power
Sources 131:86–90
14
1 Research Background
46. Waldbillig D, Kesler O (2009) Characterization of metal-supported axial injection plasma
sprayed solid oxide fuel cells with aqueous suspension plasma sprayed electrolyte layers.
J Power Sources 191:320–329
47. Nedelec R, Neagu R, Uhlenbruck S, Maric R, Sebold D, Buchkremer HP et al (2011) Gas
phase deposition of diffusion barriers for metal substrates in solid oxide fuel cells. Surf Coat
Technol 205:3999–4004
48. Hathiramani D, Vaßen R, Mertens J, Sebold D, Haanappel VAC, Stöver D (2008)
Degradation mechanism of metal supported atmospheric plasma sprayed solid oxide fuel
cells. In: Advances in solid oxide fuel cells II: ceramic engineering and science proceedings.
Wiley, New York, pp 55–65
49. Ding D, Li X, Lai SY, Gerdes K, Liu M (2014) Enhancing SOFC cathode performance by
surface modification through infiltration. Energy Environ Sci 7:552–575
50. Tucker MC, Lau GY, Jacobson CP, DeJonghe LC, Visco SJ (2007) Performance of
metal-supported SOFCs with infiltrated electrodes. J Power Sources 171:477–482
51. Blennow P, Hjelm J, Klemenso T, Ramousse S, Kromp A, Leonide A et al (2011)
Manufacturing and characterization of metal-supported solid oxide fuel cells. J Power Sources
196:7117–7125
52. Joo JH, Choi GM (2008) Simple fabrication of micro-solid oxide fuel cell supported on metal
substrate. J Power Sources 182:589–593
53. Kim KH, Park YM, Kim H (2010) Fabrication and evaluation of the thin NiFe supported solid
oxide fuel cell by co-firing method. Energy 35:5385–5390
54. Park YM, Kim JH, Kim H (2012) Effects of a current treatment for an in-situ sintered cathode
in a Ni-supported solid oxide fuel cell. Int J Hydrogen Energy 37:555–565
55. Nielsen J, Hjalmarsson P, Hansen MH, Blennow P (2014) Effect of low temperature in-situ
sintering on the impedance and the performance of intermediate temperature solid oxide fuel
cell cathodes. J Power Sources 245:418–428
56. Kim Y-M, Bae J (2009) Investigation of mixed conducting cathode for metal-supported
SOFC. ASME Conf Proc 2009:847–849
57. Baek S-W, Jeong J, Kim Y-M, Kim JH, Shin S, Bae J (2011) Metal-supported solid oxide fuel
cells with barium-containing in-situ cathodes. Solid State Ionics 192:387–393
Chapter 2
Fabrication and Investigation
of Intermediate-Temperature MS–SOFCs
2.1
Introduction
Metal-supported solid oxide fuel cells (MS–SOFCs) offer many advantages like
excellent structural robustness and stability, high tolerance toward rapid thermal
cycling, easy stack assembling as well as low materials cost over the conventional
all-ceramic structured solid oxide fuel cells (SOFCs) [1–3]. Such advantages benefit
from the mechanically robust, electrically and thermal conductive metals applied in
the metal-supported construction. However, employing metals, e.g., stainless steel
as the support may bring a number of challenges to the cell fabrication processes,
especially the densification of electrolytes and the fabrication of electrodes.
Densification of electrolyte materials like yttria stabilized zirconia (YSZ)
requires high temperatures (usually above 1200 °C). However, serious oxidation of
the metal substrate will happen at the high sintering temperatures. Co-firing the
metal support and electrolyte in a reducing atmosphere can solve the problem, while
electrolytes like doped ceria and strontium and magnesium doped lanthanum gallate
(LSGM) are not stable under the reducing atmosphere at high temperatures.
Meanwhile, elements inter-diffusion between the ferritic FeCr substrate and
nickel-containing anode, e.g., iron and chromium from the ferritic substrate into the
anode and nickel from the anode into the ferrite is a serious problem resulting in the
low power density and rapid performance degradation of MS–SOFCs [4, 5].
Furthermore, fabrication of the cathode layer is also challenging for MS–SOFCs
since the commonly used cathode materials such as La1−xSrxMnO3−d (LSM) and
La1−xSrxCo1−yFeyO3−d (LSCF) would decompose when sintered at high temperatures in a reducing atmosphere that is required to protect the stainless steel
substrates from excessive oxidation. Such processing challenges make the electrochemical performance and stability of the most MS–SOFCs much lower than
those of the traditional all-ceramic structured SOFCs. Even though techniques like
plasma spray and pulsed laser deposition (PLD) are applied to the MS–SOFCs
© Springer Nature Singapore Pte Ltd. 2018
Y. Zhou, Study on Fabrication and Performance of Metal-Supported Solid Oxide
Fuel Cells, Springer Theses, https://doi.org/10.1007/978-981-10-6617-7_2
15
16
2
Fabrication and Investigation of Intermediate-Temperature …
preparation processes, such methods would either increase the fabrication costs or
sacrifice the cell performances [6–11].
Due to the above issues, most state of the art MS–SOFCs operate at temperatures
around 800 °C, these high temperatures may cause problems like substrate oxidation
and performance degradation. How to reduce the operation temperature of MS–
SOFCs with YSZ electrolyte to intermediate-temperatures (600–800 °C) via a simple
and low-cost method is a real problem. In recent years, the infiltration method has
been applied into the manufacture of MS–SOFCs [1, 12–14]. This method involves
preparing a porous backbone, e.g., yttria-stabilized-zirconia (YSZ) backbone which
has been sintered at a high temperature (around 1300 °C). The second component of
the electrode is then introduced into the porous backbone by infiltrating and subsequent oxidizing or reducing at a low temperature (350–850 °C). The infiltration
method not only avoids the high-temperature process but also enables promising cell
performances for the resulting nano-structured catalysts.
Here we design a novel MS–SOFC based upon tri-layers-porous 430L substrate |
dense YSZ electrolyte | porous YSZ backbone-with Ni/Ce0.8Sm0.2O2−d (SDC)/
Ni–SDC catalysts infiltrated into the porous 430L substrate as the anode and
La0.6Sr0.4Fe0.9Sc0.1O3−d (LSFSc) catalysts infiltrated into the porous YSZ backbone
as the cathode, respectively. This simplified tri-layer-structure not only reduces the
cell manufacturing processes but also eliminates the resistances caused by additional barrier layers. Moreover, since the active electrode catalysts are deposited
into the pre-sintered backbones at relatively low temperatures, problems like elemental inter-diffusion and stainless steel oxidation could be avoided.
2.2
2.2.1
Experimental Section
Technical Route
The schematic of the technical route for the manufacturing of MS–SOFCs is shown
in Fig. 2.1. Firstly, green tapes based upon tri-layers-metal support (pore formers
containing) | YSZ electrolyte | YSZ electrolyte (pore formers containing) were
fabricated by the tape casting and laminating techniques. Secondly, the green tapes
were co-sintered at a high temperature (around 1300 °C) under the reducing
atmosphere and a structure of “porous metal support | dense YSZ electrolyte |
porous YSZ” was obtained. Thirdly, precursor solutions of the anode and cathode
materials were infiltrated into the porous metal support and the porous YSZ layer,
respectively. Lastly, a low-temperature heat treatment (350–850 °C) was conducted
to convert the precursors into nano particles acting as active electrode materials for
the MS–SOFCs.
2.2 Experimental Section
17
Fig. 2.1 Schematic of the production process for the manufacturing of MS–SOFCs
2.2.2
Fabrication of Symmetric and Single Cells
For single cell preparation, commercial 430L stainless steel powder (−400 mesh,
Jing-yuan Powder Material Co., Ltd, China) and 8YSZ powder (Tosoh
Corporation, Japan) were used as starting materials. The slurry for tape casting was
ethanol based which contained pore-forming agent, dispersing agent, binder,
plasticizer and other organic additives, in addition to powders. The simple tri-layer
structure of porous 430L | YSZ electrolyte | porous YSZ backbone was produced by
laminating tape cast green tapes and subsequent co-firing at 1300 °C for 4 h in a
reducing atmosphere (5% H2/95% N2). Symmetric anode and cathode cells were
prepared similarly, based upon “porous 430L | YSZ electrolyte | porous 430L” and
“porous YSZ | YSZ electrolyte | porous YSZ”, respectively. Both the symmetric
anode and cathode cells were supported by a dense YSZ electrolyte with the
thickness of 200 µm. For the symmetric anode cell, the thickness of the porous
430L was 80 µm. While for the symmetric cathode cell, the thickness of the porous
YSZ was 40 µm.
For cathode catalysts, LSFSc particles were introduced into the porous YSZ
backbones by infiltrating aqueous solutions containing stoichiometric amounts of La
(NO3)3, Sr(NO3)2, Fe(NO3)3 and Sc(NO3)3, where citric acid was also added at a 1:1
molar ratio to metal ions. After drying, heat treatment was conducted at 850 °C in a
reducing atmosphere of 5% H2–95% N2 for 2 h to convert these salts into metal
oxides without excessive oxidation of the 430L substrate. While for the anode, Ni
(NO3)2, Sm(NO3)3 and Ce(NO3)3 aqueous solution in stoichiometric ratios (the mass
ration of SDC:Ni = 8:2/1:0/0:1) was also introduced into the porous 430L support
by the infiltration method after the cathode preparation. Heat treatment of the anode
catalysts was conducted at 600 °C. The loadings of infiltrated catalysts were controlled by a micro-liter syringe each time and the infiltration/heat treating cycle was
18
2
Fabrication and Investigation of Intermediate-Temperature …
repeated to achieve the ultimate loadings needed. A single infiltration/heat treating
cycle yielded the loading of 5 wt% for cathode and 3 wt% for anode.
2.2.3
Material Characterizations
The phases of infiltrated electrodes were identified by a Rigaku XRD diffractometer
at room temperature, using monochromatic CuKa radiation. The microstructures of
cells were studied using scanning electron microscopy (SEM) in S-4800-II
microscopes. The porosities and pore-size distributions of porous 430L substrates
and YSZ backbones were measured using mercury intrusion porosimetry carried
out with a Micromeritics Auto Pore IV 9500 V1.09. Atomic diffusions were analyzed using a JEOL JXA-8100 electron probe microanalyzer (EPMA). In situ X-ray
diffraction was performed on a Bruker–D8 advanced X-ray diffractometer equipped
with an Anton Paar HTK1200 using Cu Ka radiation source. During the experiments, the as-synthesized powders were annealed in air and 5% H2–95% Ar at an
interval of 50 °C between 600 and 750 °C. The diffraction patterns were collected
with a step of 0.02o over the range of 20–80o. The reduction behavior of assynthesized SDC powders was studied by hydrogen temperature-programmed
reduction (H2–TPR) using Micromeritics ChemiSorb 2720 instrument. Specifically,
20 mg of powders were packed in a U-type quartz reactor and degassed in He at
350 °C for 0.5 h, followed by cooling to 25 °C under He flow. The H2–TPR profile
was collected in a stream of 5% H2–95% Ar (30 sccm) at a ramp rate of 5 °C/min
up to 1000 °C. The effluent gas was analyzed by a thermal conductivity detector
(TCD).
2.2.4
Electrochemical Measurements
For electrochemical measurements, single fuel cells were sealed onto alumina tubes
using silver paste (DAD–87, Shanghai Research Institute of Synthetic Resin).
Silver grids were applied onto electrodes as current collectors with silver wires
attached as the voltage and current leads. Current–voltage curves were obtained
using an IM6 Electrochemical Workstation (ZAHNER, Germany) at 600–800 °C
with cathodes exposed to air and anodes to humidified (3% H2O) hydrogen both at
100 sccm. Electrochemical impedance spectra (EIS) were collected at open circuits
with 20 mV AC amplitudes over the frequency range of 0.02 Hz–0.2 MHz.
Impedance measurements were also performed in air on symmetric cathode cells or
in humidified (3% H2O) hydrogen on symmetric anode cells. Active areas of the
single cell, symmetric anode cell and symmetric cathode cell were 0.35, 0.7 and
0.35 cm2, respectively.
2.3 Results and Discussion
2.3
2.3.1
19
Results and Discussion
Investigation of Infiltrated LSFSc–YSZ Cathodes
Figure 2.2 shows the XRD patterns of the infiltrated LSFSc–YSZ cathode as
obtained with subsequent oxidation at 800 °C for 2 h in air and formation of the
LSFSc perovskite oxide is confirmed [15].
The polarization resistances of the infiltrated LSFSc–YSZ composite cathodes
measured at 800 °C in air with the LSFSc loadings ranging from 20 to 40 wt% are
shown in Fig. 2.3a. It can be found that the composite cathode with the LSFSc
loading of 30 wt% exhibits the lowest polarization resistance of 0.024 X cm2. Such
result is lower than the LSCF or LSF infiltrated YSZ cathode as reported before [16,
17]. Note that both the polarization resistances of LSFSc loadings lower and higher
than 30 wt% are much larger than that of the 30 wt% loading. It can be explained by
the three-phase boundary (TPB) dependence of the infiltrated loadings. Model
study indicates that the total and active TPB lengths initially increase with
increasing infiltration loading, reach a maximum value, and then decrease with a
further increase in infiltration loading due to the contact between the infiltrated nano
particles [18]. EIS of infiltrated LSFSc–YSZ cathodes with different LSFSc loadings in Fig. 2.3b, c indicate the reducing of polarization resistance (mainly in the
intermediate- and high-frequency range) when the LSFSc loading increases from 20
to 30 wt%. While further increasing the loading to 35 and 40 wt%, enlarged
intermediate- and high-frequency arcs are shown. Since the intermediate-frequency
arcs are frequently attributed to the oxygen surface reaction and high-frequency arcs
to the charge transfer process [19], one possible explanation is that when the
loading is lower than 30 wt%, infiltrated particles are not sufficient to build enough
surface areas for surface oxygen adsorption and enough pathways for electron
transfer, while when the loading is higher than 30 wt%, the aggregated LSFSc
particles would reduce the effective TPB lengths of the LSFSc–YSZ composite
electrodes. In contrast to the changeable intermediate- and high-frequency arcs, the
small low frequency arcs attributed to the process of gas diffusion are relatively
Fig. 2.2 XRD patterns
of the LSFSc–YSZ cathode
with subsequent oxidation
at 800 °C in air. Reproduced
with permission from Ref.
[15]. Copyright 2014, The
Electrochemical Society
20
2
Fabrication and Investigation of Intermediate-Temperature …
Fig. 2.3 a Polarization resistances, b Bode representations of EIS, c Nyquist representations of
EIS and d Activation energies of the polarization resistances of the LSFSc–YSZ cathodes with
different LSFSc loadings. Reproduced with permission from Ref. [15]. Copyright 2014, The
Electrochemical Society
stable as observed in Fig. 2.3b, c. We surmise that the LSFSc loadings within the
range of 20–40 wt% may not affect the gas diffusion process considering the high
porosity of YSZ backbone (69%). Figure 2.3d shows the activation energies of the
polarization resistances of the LSFSc–YSZ cathodes with different LSFSc loadings.
As observed, the 30 wt% loading shows the lowest activation energy, e.g., 1.10 eV
for 30 wt% while 1.18 eV for 25 wt% and 1.19 eV for 35 wt%.
In the cathode, oxygen is reduced to oxide ion via the following oxygen
reduction reaction (ORR):
1
O ðgÞ þ 2e
2 2
¼ O2
ð2:1Þ
In Kroger–Vink notation, it is:
1
O ðgÞ þ 2e þ VO::
2 2
¼ O
O
ð2:2Þ
2.3 Results and Discussion
21
ORR is a complex multi-step reaction, including: the gas diffusion, adsorption
and dissociation of oxygen on the surface of the cathode, and the charge transfer,
etc. Specifically:
Diffusion and adsorption of the oxygen on the surface of the cathode:
O2 ðgÞ , O2ðadÞ
ð2:3Þ
Dissociation of the adsorbed molecular oxygen into atomic oxygen:
O2ðadÞ , 2Oad
ð2:4Þ
Charge transfer from the cathode to the atomic oxygen (reduction of the
adsorbed atomic oxygen):
Oad þ e , O
ad
ð2:5Þ
Solid-phase migration of the oxide ion from the cathode to the TPB:
O
ad , OTPB
ð2:6Þ
Reduction of the monovalent oxide ion in the TPB:
2
O
TPB þ e , OTPB
ð2:7Þ
Diffusion of the oxygen ion from TPB to the electrolyte, associated with a charge
transfer process:
::
O2
TPB þ VO , OO
ð2:8Þ
The widely used parameter to determine the rate-determining step of the ORR is
the slope of the polarization resistance of the cathode as a function of oxygen partial
pressure, following the relation:
RP / Pn
O2
ð2:9Þ
When the value of n is 1, 1/2, 3/8, 1/4, 1/8 and 0, the corresponding
rate-determining step of the ORR is (2.3), (2.4), (2.5), (2.6), (2.7) and (2.8),
respectively [20–25]. Figure 2.4 shows the relations of polarization resistance to
PO2, and schematics of their corresponding ORR processes [26].
In order to gain insights into oxygen reduction kinetics and identify the
rate-limiting step of the infiltrated LSFSc–YSZ cathode, impedance data were
collected for the symmetric cathode cells with 30 wt% of LSFSc at varied temperatures and under varied oxygen partial pressures. These impedance data were
further fitted with an equivalent circuit, Ro (RH, CPEH) (RL, CPEL), where Ro was
the pure ohmic resistance due to the electrolyte and the electrodes, RH and RL were
widths of the high- and low-frequency arcs, while CPEH and CPEL were the
22
2
Fabrication and Investigation of Intermediate-Temperature …
Fig. 2.4 The relations of polarization resistance to PO2, and schematics of their corresponding
ORR processes on the cathode. Reproduced with permission from Ref. [26]. Copyright 2014,
Elsevier
Fig. 2.5 Schematics of the equivalent circuit used for impedance spectra fitting
constant phase elements for the high- and low-frequency arcs, respectively
(Fig. 2.5).
Figure 2.6 shows the Nyquist plots of the polarization resistances of the LSFSc–
YSZ symmetric cells measured in air at different temperatures and oxygen partial
L
pressures. The high-frequency value (RH
PC) and low-frequency value (RPC) as a
H
function of temperature are shown in Fig. 2.7a. The RPC value followed an
Arrhenius dependence with an activation energy of 1.07 eV, while the RLPC value
remained almost unaltered. It indicates that the high-frequency arc may associate
with the charge transfer process in the interface of LSFSc and YSZ, and the
low-frequency arc may relate to the diffusion and adsorption process of oxygen.
This conclusion is further verified by the data measured under various oxygen
partial pressures. As shown in Figs. 2.6b and 2.7b, a much more obvious oxygen
partial pressure dependence of the low-frequency arcs was found than that of the
2.3 Results and Discussion
Fig. 2.6 Impedance plots of
LSFSc–YSZ symmetric
anode cells measured: a At
650–800 °C, b At 800 °C in
atmospheres with different
oxygen partial pressures
Fig. 2.7 a Temperature
dependence of the RH and RL
result, b Oxygen partial
pressure dependence of the
RH and RL result measured at
800 °C
23
24
2
Fabrication and Investigation of Intermediate-Temperature …
L
high-frequency arcs. In addition, the RH
PC value is much higher than the RPC value
in the oxygen partial pressure range of 0.21–0.1 atm, thus the ORR of the
LSFSc–YSZ cathode is limited by the charge transfer process in the interface of
LSFSc and YSZ.
2.3.2
Investigation of Infiltrated Ni–430L Anodes
and the MS–SOFCs
Figure 2.8 shows the polarization resistances of the infiltrated Ni–430L anodes
(calcining at 600 and 850 °C) measured at 650 °C in humidified hydrogen with the
Ni loadings ranging from 4 to 16 wt% [27]. For the infiltrated Ni–430L anodes
calcined at 600 °C, the Ni loading of 10 wt% exhibits the lowest polarization
resistance of 2.2 X cm2. Note that both the polarization resistances of Ni loadings
lower and higher than 10 wt% are much larger than that of the 10 wt% loading. It
can be explained by the three-phase boundary (TPB) dependence of the infiltrated
loadings. Ni loadings dependence of the polarization resistances of the Ni–430L
anodes calcined at 850 °C is also shown in Fig. 2.8. Different from the results of the
anodes calcined at 600 °C, the optimized loading is 13 wt% for the anodes calcined
at 850 °C. It can be explained that when the Ni particles are coarsened, connected
particles tend to be isolated and an increased loading is needed to provide sufficient
active surfaces for the hydrogen oxidation reaction [2].
Polarization resistances of the Ni–430L anodes obtained here are extraordinarily
large compared to those of the Ni infiltrated doped LaGaO3 anode, which only
exhibited a low polarization resistance of 0.008 X cm2 at 650 °C [28]. This is very
likely caused by the short TPB length, which is only confined to the narrow YSZ–
Ni contact area (marked in red in Fig. 2.9) closed to the YSZ electrolyte due to the
absence of oxide-ion conducting components in the infiltrated Ni–430L anodes.
Impedance spectra of the infiltrated Ni–430L anode (10 wt% loading) calcined at
600 °C is shown in Fig. 2.10a. Notably, hydrogen oxidation kinetics is largely
Fig. 2.8 Polarization
resistances of the Ni–430L
anodes with various Ni
loadings tested at 650 °C.
Reproduced with permission
from Ref. [27]. Copyright
2014, Elsevier
2.3 Results and Discussion
25
Fig. 2.9 Schematic diagram depicting the reaction pathways in Ni–430L anode
dominated by the charge transfer process with the summit relaxation frequency at
826 Hz, while the surface hydrogen exchange process is quick enough and makes
little contributions to the anode polarization resistance since the arc commonly
centered at 1 Hz is negligibly small. In order to evaluate the stability of the
polarization resistance, the symmetrical anode cell was operated at 650 °C for about
50 h without current applied. The impedance spectra of the Ni–430L anode after
the durability measurement is shown in Fig. 2.10b. Note that the anode polarization
resistance increases drastically from 2 to 41 X cm2 during the 50-h measurement
and an increase in the lower-frequency is shown. Such increase is much more
obvious in the Bode representations of the impedance data shown in Fig. 2.10c. It is
reported that the higher-frequency arc is related to the charge transfer process near
Fig. 2.10 Impedance data of the Ni–430L anodes calcined at 600 °C. Reproduced with
permission from Ref. [27]. Copyright 2014, Elsevier
26
2
Fabrication and Investigation of Intermediate-Temperature …
the TPB region, while the lower-frequency arc to the hydrogen dissociation
adsorption or surface diffusion process on the Ni surface [14, 29, 30]. Since the
infiltrated Ni particles are easy to be coarsened, the coarsening of Ni particles which
decreasing the density of TPBs and active surfaces for hydrogen oxidation reactions
may be the main reason to the increase of the anode polarization resistances [2, 19].
A sustained increase in the polarization resistance of the Ni–430L anode during the
50-h measurement is shown in Fig. 2.10d. For the large polarization resistance after
the durability test, no extended measurement was applied.
Impedance spectra of the Ni–430L anode calcined at 850 °C (13 wt% loading)
before and after the stability test are shown in Fig. 2.11a. A comparing of the
impedance arcs in Figs. 2.10a and 2.11a (initial results) shows the similar arc shape
and frequency distribution. It exhibits that coarsening of the Ni particles at 850 °C
may not change the electrochemical process in the anode. However, the stability of
Fig. 2.11 Impedance of the
Ni–430L anodes calcined at
850 °C. Reproduced with
permission from Ref. [27].
Copyright 2014, Elsevier
2.3 Results and Discussion
27
the polarization resistance of the anode is improved and an increase from 2.6 to
4.3 X cm2 during the 200-h measurement is exhibited in Fig. 2.11a. As shown in
Fig. 2.11a, b, the increase of the polarization resistance mainly centered at the
middle-frequency after the durability test. It indicates that the Ni particles may be
further coarsened after the 200-h stability measurement. Compared to the rapid
increase of the polarization resistance for the Ni–430L anode calcined at 600 °C, a
much more slowly increase in the polarization resistance is shown for the anode
calcined at 850 °C. This result illustrates that pre-coarsening of the infiltrated Ni
particles can enhance the stability of the Ni–430L anode [2].
Scanning electron microscope (SEM) images of the Ni microstructures before
and after the durability measurement are shown in Fig. 2.12. For the Ni calcined at
600 °C, pronounced particle growth is observed. In particular, well interconnected
nickel layers (Fig. 2.12a) increase to the isolated coarsening particles after the
stability test (Fig. 2.12b). Such phenomenon consists with the impedance spectra
change shown in Fig. 2.10a, b. While for the Ni particles calcined at 850 °C, even
the coarsening phenomenon is still observed (Fig. 2.12c, d), it is not so serious as
that of the particles calcined at 600 °C. That may be the reason why the polarization
resistance of the infiltrated anode calcined at 850 °C increases more slowly
with time.
Fig. 2.12 Microstructures of the infiltrated Ni particles calcined at 600 °C: a Before and b after
the stability test, microstructures of the infiltrated Ni particles calcined at 850 °C: c Before and
d after the stability test. Reproduced with permission from Ref. [27]. Copyright 2014, Elsevier
28
2
Fabrication and Investigation of Intermediate-Temperature …
Fig. 2.13 SEM images of the: a MS–SOFC, b LSFSc–YSZ cathode, c Ni–430L anode.
Reproduced with permission from Ref. [31]. Copyright 2014, Elsevier
MS–SOFCs based on the infiltrated Ni–430L anode and the LSFSc–YSZ
cathode have been fabricated. Figure 2.13a shows a representative cross-sectional
SEM micrograph of the single fuel cell with the infiltrated catalyst loadings of 10 wt
% for the Ni anodes and of 30 wt% for the LSFSc cathodes. The YSZ electrolyte
layer is fully dense with a typical thickness of 30 lm, and is well bonded with the
adjacent 430L substrates and porous YSZ backbones. The porous YSZ backbone is
45 µm thick and the porous 430L substrate is 300 µm (part of the substrate is not
shown in Fig. 2.13a). As shown in Fig. 2.13b, the infiltration of 30 wt% LSFSc is
sufficient to produce well-interconnected coatings with an average particle size of
100 nm on the porous YSZ backbones. Porous 430L substrate with infiltrated Ni
particles (10 wt% Ni loading) is also shown in Fig. 2.13c.
Figure 2.14a shows typical cell voltages and power densities as a function of
current densities for the MS–SOFC operating on humidified hydrogen fuels and air
oxidants at 650–800 °C. The open circuit voltage (OCV) is 1.06 V at 800 °C,
indicating good gas impermeability of the dense YSZ electrolyte thin films as
consistent with the SEM observations in Fig. 2.13a. The maximum power densities
(MPDs) measured are 193, 418, 636 and 907 mW cm−2 at 650, 700, 750 and 800 °C,
Fig. 2.14 Electrochemical characteristics of the single cell measured at 650–800 °C: a I–P–V
characteristics, b Impedance spectra at open circuits. Reproduced with permission from Ref. [31].
Copyright 2014, Elsevier
2.3 Results and Discussion
29
respectively. These values are highly comparable with those previously achieved for
alternative MS–SOFCs as fabricated using low-temperature deposition method like
plasma spraying [32]. For example, FeCrMnTi supported vacuum plasma sprayed
(VPS) Ni–YSZ anode and YSZ electrolyte fuel cells with suspension plasma sprayed
(SPS) LSCF–CGO cathode showed a power density of 798 mW cm−2 at 0.7 V
(800 °C) with H2/O2 atmosphere on anode/cathode side [33]. Figure 2.14b shows
Nyquist plots of impedance data as obtained at open circuits for the present
MS–SOFCs, where the pure ohmic losses (RO) were taken from the high-frequency
real-axis intercepts and the combined anode and cathode polarization resistances
(RP) corresponded to the overall width of the depressed arcs. In particular, the pure
ohmic losses are RO = 0.34, 0.22, 0.17 and 0.15 X cm2 (shown in the magnified
figure), and the total interfacial polarizations are RP = 2.04, 1.32, 1.05 and 0.85 X
cm2 at 650, 700, 750 and 800 °C, respectively. Compared to the relatively small
ohmic losses, the cell performances are largely limited by the polarization
resistances.
In order to examine the long-term stability of the nano-scale catalysts, the present MS–SOFC was operated for 200 h under a constant terminal voltage of 0.7 V
at 650 °C. Figure 2.15a shows a constant decrease of the current density from 190
to 147 mA cm−2. Figure 2.15b compares Nyquist plots of the impedance data taken
at open circuits before and after the durability test, indicating that the ohmic losses
remain nearly unchanged while the total interfacial polarization resistances increase
from 1.49 to 5.43 X cm2. The degradation rate for the present fuel cells as estimated
from Fig. 2.15a is 11%/100 h, which is much lower than that of prior MS–SOFCs
with infiltrated Ni catalysts operated at a higher temperature of 700 °C with the
power density dramatically decreasing from 250 to 50 mW cm−2 during a 15-h
measurement [2]. Nevertheless, the present degradation rate is substantially larger
than the value of 1%/1000 h as obtained at the same temperature for MS–SOFCs
with co-infiltrated Ce0.8Gd0.2O2−d-Ni anode catalysts during a 3000-h durability
measurement [34].
Fig. 2.15 a Stability of the single cell measured at 650 °C, b Impedance spectra of the single cell
measured before and after the stability test. Reproduced with permission from Ref. [31]. Copyright
2014, Elsevier
30
2
Fabrication and Investigation of Intermediate-Temperature …
As shown in Fig. 2.15b, it is the increased total interfacial polarization resistance
that yields the large drop in the cell power density. To identify the individual
contributions of the anode and the cathode to such an increase in the total polarization resistance, impedance measurements were also performed on the symmetric
cathode and the symmetric anode cells over the long term, and the results are
summarized in Fig. 2.16a, b, respectively. Notably, the anode polarization resistance increases drastically from 2 to 30 X cm2 during the 30-h measurement at 650
°C. In contrast, the cathode polarization resistance stays almost unchanged over an
even longer measurement of 400 h. Therefore, formation of insulating impurities
and coarsening of nano-scale catalysts, which are normally identified as the two
main reasons for decreased catalytic activities and increased polarization resistances
of the infiltrated cathodes, did not occur for the present LSFSc–YSZ composites at
650 °C [35, 36]. Good stability in the nano-scale LSFSc catalysts is further supported by the SEM examination of the cathode microstructure before and after the
durability measurement (Fig. 2.17a, b), where no pronounced increase in the particle size or change in the particle morphology is observed. On the other hand,
pronounced growth in the nano-scale Ni catalysts is observed for the Ni–430L
composites. In particular, well interconnected nickel particles with an average
particle size of 0.1–0.3 µm evolve into isolated and coarsened ones with particle
size of 0.5–1 µm after the stability test (Fig. 2.17c, d). The morphological evolution
of the Ni particles decreases the density of TPBs for hydrogen oxidation reactions
and thereby increases the anode polarization resistances as shown in Fig. 2.16b.
Prior reports have shown that the metallic inter-diffusion between metal substrates and Ni based anodes may occur during the fabrication and operation of fuel
cells, decreasing the anode catalytic activities and thus increasing the polarization
resistances of the anodes [37, 38]. For example, diffusion depth of Fe and Cr in the
nickel anodes was as large as 200 µm when heat-treated at 1400 °C for 2 h in 4%
H2–96% Ar [39]. In order to identify whether inter-diffusions of Fe, Ni and Cr
occurred at the operating temperature of MS–SOFCs, porous Ni coatings were
Fig. 2.16 Stabilities of the polarization resistances of the symmetric cells: a cathode, b anode.
Reproduced with permission from Ref. [31]. Copyright 2014, Elsevier
2.3 Results and Discussion
31
Fig. 2.17 Microstructures of the LSFSc particles: a Before and b after the stability test,
microstructures of the Ni particles: c Before and d after the stability test. Reproduced with
permission from Ref. [31]. Copyright 2014, Elsevier
electroplated onto the surface of commercial 430L plates and the resulting samples
were thermally treated in 97% H2–3% H2O at 650 °C for 230 h. SEM examination
shows that the Ni coating is well adhered onto the 430L substrate, and EDX
analysis indicates that the Ni, Cr and Fe concentrations decrease gradually along the
interfacial region of 30 µm thick between Ni and 430L (Fig. 2.18), which is
Fig. 2.18 Microstructure and
elements distributions at the
interface of Ni coating (left)
and 430L background (right)
after the heat-treatment at
650 °C for 230 h: Ni-red, Fegreen, Cr-blue. Reproduced
with permission from Ref.
[31]. Copyright 2014,
Elsevier
32
2
Fabrication and Investigation of Intermediate-Temperature …
probably caused by the inter-diffusions of Fe, Cr and Ni between the deposited Ni
layer and the 430L substrate over the measurement period. Therefore, atomic diffusion could also be one of the possible physical mechanisms for the cell degradation in addition to the Ni coarsening.
Other than Ni coarsening and metallic inter-diffusion, oxidation of the porous
430L substrates in humidified hydrogen could also increase the anode area specific
resistance due to the formation of oxide scales and the metal/oxide scale interfaces
[40, 41], especially given that the present 430L substrates have increased surface
area available for oxidation than the dense counterpart [42]. Note that the ohmic
losses remained constant during the long-term measurements for the functioning
single cells (Fig. 2.15b), indicating that the resulting scales had sufficiently high
conductivities or the oxidation itself was not so serious at all. Furthermore, there is
no obvious spalling observed between the Ni coatings and the 430L substrates as
shown in Fig. 2.17d. Therefore, based upon the present preliminary durability
measurement, oxidation of the 430L substrates might not be a significant issue for
the present fuel cells. This conclusion is consistent with previous reports that 430L
can achieve the oxidation rates required for the fuel cell lifetime of 50,000 h when
the operating temperature is within the range of 650–700 °C [1].
2.3.3
Investigation of Infiltrated SDC–430L Anodes
and the MS–SOFCs
The infiltrated Ni–430L anode showed a good electrochemical performance when
applied in MS–SOFCs. However, the application of such anodes is hindered by the
performance degradation caused by the Ni coarsening and metallic inter-diffusion
issues. Thus, a ceramic anode with a good morphological and chemical stability
should be taken into consideration. Ceria and ceria-based oxides are attractive as the
SOFC anode catalysts due to their excellent redox and catalytic properties, as
enabled by the distinctive feature of cerium ions to easily switch between Ce4+ and
Ce3+ in different atmospheres [43–46]. In this section, Ce0.8Sm0.2O2−d (SDC) is
applied as the infiltrated anode catalysts and the particle coarsening and metallic
inter-diffusion issues can be avoided, thus, the high long-term stability is expected.
The SDC powder was synthesized by calcining the infiltration solution in air at
800 °C for 2 h. Cubic particles with a particle size of 40–140 nm are shown in
Fig. 2.19.
The interaction of SDC nano-particles with H2 was examined by performing
temperature-programmed reduction (TPR) in a 5% H2–Ar gas flow, with a typical
TPR profile shown in Fig. 2.20a. Two hydrogen consumption peaks, one at 633 °C
and the other at 932 °C are observed due to different oxygen binding energies in the
fluorite lattice. The peak at low temperatures is usually attributed to the formation
of surface hydroxyl and removal of surface oxygen (i.e., reduction of surface
cerium ions), whereas the peak at high temperatures is normally due to bulk
2.3 Results and Discussion
33
Fig. 2.19 Microstructure of
the SDC powder as
synthesized
reduction of Ce4+ to Ce3+ [47]. Nearly the same intensity for these two peaks
suggests that nano-scale SDC catalysts have significant amounts of surface oxygen
to be extracted by hydrogen. This fact can be well understood by large
area-to-volume ratios of nano-scale SDC particles and high mobility of oxygen
anions in the SDC lattice that would make bulk oxygen more readily available for
surface extraction via diffusive jumps of free oxygen vacancies. The total amount of
the releasable surface oxygen is 0.315 lmol per gram of SDC as estimated from the
TRP profile in Fig. 2.20a.
The reduction behavior of SDC nano-particles in Fig. 2.20a was also reflected in
the evolution of their crystal structure. In situ high temperature X-ray diffraction
(XRD) measurements indicate that the cubic fluorite crystal structure was well
maintained in H2 (Fig. 2.21). Temperature-dependent evolution of the reflection
peak (111) is shown in Fig. 2.20b, c for SDC oxides in hydrogen and air,
respectively. Figure 2.20d summarizes the lattice parameter at different temperatures, as calculated from the Bragg angle of (111) peak in Fig. 2.20b, c. In contrast
to a linear increase of the lattice parameter in air with increasing temperature, an
abrupt expansion is observed in hydrogen at temperatures above 600 °C. This
phenomenon should be related to irreversible association of adjacent surface
hydroxyl to release gaseous water molecules that results in a reduction of cerium
ions from Ce4+ to Ce3+ and a concomitant chemical expansion of the lattice due to
larger radius for Ce3+ (0.114 nm) than for Ce4+ (0.097 nm) [48]. Such a lattice
expansion behavior well coincides with the low-temperature hydrogen consumption
peak with Tm = 633 °C in the TPR profile shown in Fig. 2.20a.
The sintered 430L scaffolds, as examined using the SEM, show a macroporous
structure with the pore size distribution ranged widely from 4 to 90 lm (Fig. 2.22a,
b). Mercury porosimetry measurements show an open porosity of 44% with pore
sizes largely centered at 17 lm. Then, the porous 430L scaffolds were immersed
into aqueous solutions containing Sm(NO3)3, Ce(NO3)3 and citric acid to produce
thin SDC coatings after calcining in hydrogen at 800 °C. Figure 2.22c shows a
34
2
Fabrication and Investigation of Intermediate-Temperature …
Fig. 2.20 Reduction behavior of the SDC catalysts: a TPR profile, Evolution of the (111) peak at
elevated temperatures for XRD patterns in b 5% H2–95% Ar and c air, d Temperature evolution of
the lattice parameter as calculated (111) peaks shown in b and c
typical cross-sectional SEM image of the resultant composites, indicating that
nanoporous SDC layers are homogeneously coated inside the macropores of 430L
scaffolds. Close examination of the deposited SDC coatings (Fig. 2.22d) reveals a
typical “cauliflower” structure, where the spherical agglomerates sizing from 60 to
110 nm appeared to consist of even finer particles of 10–20 nm in diameter. Upon
increasing the calcining temperature to 1200 °C, these nanoscale particles grew
dramatically and spread out on the 430L scaffolds to form dense thin films. SEM
images for dense SDC–430L composites (Fig. 2.22e, f) show that such dense thin
films consist of grains of 0.25–0.5 lm. Nonetheless, large surface-to-volume ratios
of nanoporous SDC–430L composites would enable facile extraction of surface
oxygen or reduction of surface cerium ions.
2.3 Results and Discussion
35
Fig. 2.21 XRD patterns of the SDC powder in: a Diluted H2, b Air
Fig. 2.22 Cross-sectional SEM images showing: a and b Blank 430L scaffold, c Nanoporous
SDC–430L composites, d Nanoporous SDC catalysts, e Dense SDC–430L composite, f Dense
SDC coatings. Reproduced with permission from Ref. [49]. Copyright 2014, Wiley-VCH
36
2
Fabrication and Investigation of Intermediate-Temperature …
Fig. 2.23 SEM images of the surfaces of 430L with different SDC loadings: a 4 wt%, b 10 wt%,
c 12 wt%
To optimize the performance of the SDC–430L anode, the relationship between
the SDC loadings and the morphology, pore structure and polarization resistance of
the anode was investigated. Figure 2.23 shows the morphology of the infiltrated
SDC particles on the surface of the 430L backbone. At a loading of 4 wt%, only
isolated SDC particles with size between 20 and 150 nm are shown (Fig. 2.23a). The
isolated SDC particles further convert into nanoporous films when the loading
increased to 10 wt% (Fig. 2.23b). However, further increasing the loading to 12 wt%
resulted in dense SDC films with aggregated particles (Fig. 2.23c).
Figure 2.24 shows the pore size distributions of the 430L substrate before and
after infiltrating of the SDC catalysts. It shows that porosities of the 430L substrate
Fig. 2.24 Pore size distributions of the 430L substrate: a Blank 430L, b With a SDC loading of 4
wt%, c With a SDC loading of 10 wt%, d With a SDC loading of 12 wt%
2.3 Results and Discussion
37
Fig. 2.25 Effect of SDC
loadings on the polarization
resistances of the SDC–430L
symmetric anode cells
are 44.0, 34.5, 29.7 and 28.7%, and the main pore sizes are 17, 11, 14 and 6(11) µm
when the loading of SDC are 0, 4, 10 and 12 wt%, respectively. A reduction in both
the porosity and the pore size are shown as the loading of the SDC catalysts
increased.
Figure 2.25 shows the polarization resistances of the infiltrated SDC–430L
anodes with SDC loadings ranging from 4 to 12 wt%. Infiltrated SDC–430L anodes
with a SDC loading of 10 wt% exhibits the lowest polarization resistance at all
temperatures between 600–800 °C, which is reasonable given that reducing the
catalyst loading would result in less active sites available for the hydrogen oxidation reaction, whereas increasing the catalyst loading would decrease the overall
porosities and the TPB length in the anode (Figs. 2.23 and 2.24).
EIS of the polarization resistances of the infiltrated SDC–430L anodes with
various SDC loadings are shown in Fig. 2.26. It is found that the difference of the
EIS are mainly centered at a low-frequency range around 1 Hz. Since the
low-frequency arc is mainly related with the surface process, it can be concluded
that both the SDC loadings higher and lower than 10 wt% is not beneficial for the
adsorption and dissociation of hydrogen on the surface of the SDC–430L anode.
As shown in Fig. 2.25, the polarization resistances of the infiltrated SDC–430L
anodes (10 wt% loading) are 0.10 ± 0.01, 0.12 ± 0.02, 0.18 ± 0.03, 0.29 ± 0.06
and 0.57 ± 0.11 X cm2 at 800, 750, 700, 650 and 600 °C, respectively. These
values are competitive when compared with the polarization resistances reported for
the common SOFC anodes, e.g., >0.15 X cm2 at 750 °C for the traditional Ni–YSZ
cermet anode [50], 0.26 X cm2 at 900 °C for La0.8Sr0.2Cr0.5Mn0.5O3−d perovskite
oxide [51], 0.27 X cm2 at 800 °C for Sr2Fe1.5Mo0.5O6−d double perovskite oxide
[52], and 0.26 X cm2 at 700 °C for the Pd-promoted CeO2−d infiltrated YSZ anode
[53]. In contrast to the high resistance of the infiltrated Ni–430L anodes (2.2 X cm2
at 650 °C, Fig. 2.8), an obvious decrease in the polarization resistance is shown.
This should be caused by the enlarged TPBs from the narrow YSZ–Ni contact area
to the whole surface of the infiltrated SDC particles (marked in red in Fig. 2.27) due
to the good ionic and electronic conductivity of SDC under the reducing
atmosphere.
38
2
Fabrication and Investigation of Intermediate-Temperature …
Fig. 2.26 EIS of the SDC–430L anodes with different SDC loadings: a Nyquist plots measured at
800 °C, b Bode plots measured at 800 °C, c Nyquist plots measured at 600 °C, d Bode plots
measured at 600 °C
Fig. 2.27 Schematic diagram
depicting the reaction
pathways in SDC–430L
anode
The overall anode reaction can be written in Kroger–Vink notation as follows:
0
H2 ðgÞ þ O
O ðYSZÞ ! H2 OðgÞ þ VO€ ðYSZÞ þ 2e ð430LÞ
ð2:10Þ
Where O
O ðYSZÞ and VO€ ðYSZÞ represent oxygen ions and oxygen vacancies in
the YSZ lattice, respectively. Both quantum chemical molecular dynamic simulation and in situ surface studies revealed formation of surface hydroxyl (OH ) and
0
O
reduction of Ce4+(CeCe ) for ceria in H2 [54, 55], indicating that the global anode
reaction on the infiltrated SDC–430L anodes might proceed in multiple consecutive
or parallel steps as follows:
2.3 Results and Discussion
39
Dissociative adsorption of hydrogen molecules:
H2 ðgÞ þ 2O
O ðSDCÞ ! 2OHO ðSDCÞ
ð2:11Þ
Surface reduction of cerium ions:
0
OHO ðSDCÞ þ Ce
Ce ðSDCÞ ! OH ðSDCÞ þ CeCe ðSDCÞ
O
ð2:12Þ
Desorption of water molecules via association of adjacent surface hydroxyl:
2OH ðSDCÞ ! H2 OðgÞ þ VO€ ðSDCÞ þ O
O ðSDCÞ
O
ð2:13Þ
Electron transport within SDC coatings and transfer to 430L:
0
0
CeCe ðSDCÞ ! Ce
Ce ðSDCÞ þ e ð430LÞ
ð2:14Þ
Transport of oxygen vacancies within SDC coatings and transfer to YSZ
electrolytes:
O
O ðYSZÞ þ VO€ ðSDCÞ ! OO ðSDCÞ þ VO€ ðYSZÞ
ð2:15Þ
Among these elementary steps, Reaction (2.14) and (2.15) are related to charge
transport properties of the infiltrated SDC coatings since they transfer electrons and
oxygen vacancies to 430L scaffolds and YSZ electrolytes, respectively. In contrast,
Reaction (2.11), (2.12) and (2.13) are surface-related with the net result of removing
surface lattice oxygen and reducing surface cerium ions from Ce4+ to Ce3+.
Figure 2.28a and b shows the Nyquist plots of the EIS data of the infiltrated SDC–
430L anodes (SDC loading = 10 wt%) calcinated at 800 and 1200 °C, respectively.
The two plots consist of two depressed arcs centered at 100 and 1 Hz,
respectively. Evolution of infiltrated SDC coatings from nanoporous to dense yields
a 5-fold increase in the high-frequency arc (RH) and a 22-fold increase in the
low-frequency arc (RL). This observation, in combination with much stronger
dependence of RL values on hydrogen partial pressures (Fig. 2.29), suggests that the
more surface-sensitive RL value probably reflect extraction of surface lattice oxygen
by hydrogen-Reaction (2.11), (2.12) and (2.13), whereas the less surface-sensitive
RH value is largely dictated by charge transport behavior of oxide-ions and electrons within the SDC coatings-Reaction (2.14) and (2.15). As a matter of fact, the
activation energies for RH and RL values are essentially unaffected by the morphology of SDC coatings within the experimental uncertainty, i.e., 0.78–0.82 eV
for RH and 0.48–0.54 for RL (Fig. 2.30). Comparing the activation energy for RH
with those for oxide-ionic conduction (0.80 eV) and for electronic conduction
(1.3–2.3 eV) in ceria-based oxides implies that the RH value is likely determined by
transport of oxide-ions within the SDC coatings-Reaction (2.15) [56]. Prior
mechanistic studies of hydrogen electro-oxidation on dense and patterned undoped
ceria anodes have shown that Reaction (2.11) and (2.12) are kinetically fast and stay
40
2
Fabrication and Investigation of Intermediate-Temperature …
Fig. 2.28 Impedance spectra
of the symmetric SDC–430L
anode cells: a Nyquist plot for
nano- porous SDC–430L
anode, b Nyquist plot for
dense SDC–430L anode,
c Polarization resistance
values for both anodes plotted
versus inverse temperatures.
Reproduced with permission
from Ref. [49]. Copyright
2014, Wiley-VCH
in equilibrium while Reaction (2.13) is rate-limiting [55]. Therefore, the RL value is
more specifically dictated by desorption of water molecules via association of
adjacent surface hydroxyl, Reaction (2.13). Several times larger RL values than RH
for dense SDC infiltrated 430L anodes (Fig. 2.30) indicate that their hydrogen
electro-oxidation kinetics is always dominated by surface desorption of water
molecules. Nevertheless, the situation is quite different for nanoporous SDC infiltrated 430L anodes due to larger surface area to volume ratios, where hydrogen
electro-oxidation is co-limited at high temperatures by surface desorption of water
molecules and bulk transport of oxide-ions (Fig. 2.28a), with the latter becoming
more important at lower temperatures due to larger activation energies (Fig. 2.30).
The anode catalytic activities of the infiltrated nanoporous SDC–430L anodes
were further examined in the MS–SOFCs. Figure 2.31 shows the cross sectional
SEM images of the MS–SOFC with SDC infiltrated in the 430L substrate (10 wt%)
2.3 Results and Discussion
41
Fig. 2.29 a Nyquist plots of
impedance data for the dense
SDC–430L anode measured
in various hydrogen partial
pressures, b Hydrogen partial
pressure dependence of the
resistance values for the highand low-frequency arcs.
Reproduced with permission
from Ref. [49]. Copyright
2014, Wiley-VCH
Fig. 2.30 The higher- and
lower-frequency arcs of the
impedance data for
nanoporous SDC–430L and
dense SDC–430L plotted
versus inverse temperature.
Reproduced with permission
from Ref. [49]. Copyright
2014, Wiley-VCH
and LSFSc infiltrated in the YSZ backbone (30 wt%). Thickness for the dense YSZ
electrolyte thin film is typically 25 lm. Nano porous SDC and LSFSc particles are
found to be well attached with the porous backbones.
Electrochemical measurements were performed on the MS–SOFCs with 3%
humidified hydrogen fuels and dry air oxidants at 650–800 °C, and Fig. 2.32a
shows typical cell voltages and power densities as a function of current densities.
The open circuit voltages range between 1.09 V at 650 °C and 1.04 V at 800 °C,
and are within 50 mV of the thermodynamically expected Nernst potentials.
42
2
Fabrication and Investigation of Intermediate-Temperature …
Fig. 2.31 SEM images of the: a MS–SOFC, b SDC–430L anode, c LSFSc–YSZ cathode
Fig. 2.32 Electrochemical characteristics of the single cell measured at 650–800 °C: a I–P–V
characteristics, b Impedance spectra. Reproduced with permission from Ref. [49]. Copyright 2014,
Wiley-VCH
Maximum power densities measured are 0.45, 0.55, 0.66 and 0.94 W cm−2 at 650,
700, 750 and 800 °C, respectively. Nyquist plots of the impedance data measured at
open circuits (Fig. 2.32b) show that the total area specific resistances are 0.239,
0.340, 0.509 and 0.851 X cm2 and the ohmic losses (RO) are 0.073, 0.097, 0.133
2.3 Results and Discussion
43
Fig. 2.33 Stability of the
single cell measured
at 650 °C
and 0.202 X cm2 at 800, 750, 700 and 650 °C, respectively. It can be found that the
resistances of the single cell mainly dominated by the polarization resistances
deriving from the electrodes. Therefore, the performance of the present MS–SOFCs
has potentials for further improvement by optimizing the structure or material of the
electrodes.
Stability of the MS–SOFC measured at 650 °C and 0.7 V is shown in Fig. 2.33
and no obvious degradation is found. Compared with the sustaining degradation
shown in Fig. 2.15a, MS–SOFCs using infiltrated SDC–430L as the anode shows a
much higher stability than that of the MS–SOFC using infiltrated Ni–430L as the
anode. The stable output power shown in Fig. 2.33 confirms that the particle
coarsening and metallic inter-diffusion issues have been well addressed by using
SDC as the anode catalyst.
2.3.4
Investigation of Infiltrated Ni–SDC–430L Anodes
and the MS–SOFCs
In this section, taking advantages of the high catalytic activity of infiltrated Ni–
430L anodes and good stability of the SDC–430L anodes, a Ni–SDC–430L anode
was developed. The weight ratio of SDC to Ni is chosen to be 8:2. We surmise that
the excessive SDC ceramic phase can restrict the growth of the Ni particles and an
enhanced stability would be obtained.
Figure 2.34a reveals the XRD patterns of the Ni–SDC infiltrated 430L anode as
obtained. It is found that after treating at 600 °C for 2 h in a reducing atmosphere
(5% H2/95% N2), pure phase of SDC can be obtained. For the low Ni loading and
the main peak overlap of Ni and 430L, Ni phase is not obvious in the XRD patterns.
Figure 2.34b shows the polarization resistances of the infiltrated Ni–SDC–430L
anode measured at 650–800 °C in 97% H2–3% H2O. The polarization resistances
are 0.075, 0.081, 0.09, and 0.112 X cm2 at 800, 750, 700 and 650 °C, respectively.
44
2
Fabrication and Investigation of Intermediate-Temperature …
Fig. 2.34 a X-Ray diffraction patterns of the Ni–SDC–430L anode, b Impedance spectra of the
symmetric Ni–SDC–430L anode cell. Reproduced with permission from Ref. [15]. Copyright
2014, The Electrochemical Society
Such results are comparable with a Ni–CGO infiltrated FeCr–YSZ cermet anode
used in another MS–SOFC, which showed a polarization resistance of 0.12 X cm2
at 650 °C [14]. It is interesting that the polarization resistances of the anode in this
study change small with temperatures, corresponding to a low activation energy of
0.31 eV. The impedance spectra shown in Fig. 2.34b are composed of small
high-frequency arcs and large low-frequency arcs at all temperatures ranging from
650 to 800 °C. It is reported that the high-frequency arc related to the charge
transfer process near the TPB region is strongly dependent on temperature, while
the low-frequency arc to the hydrogen dissociation adsorption or surface diffusion
process on the anode surface is independent on temperature. Since the whole
impedance spectra of the Ni–SDC–430L anode obtained here is dominated by the
large low frequency arc (Fig. 2.34b), it is no wonder that the activation energy of
the anode is very low. Above all, the low activation energy indicates that the Ni–
SDC–430L anode is appropriate to be operated at low temperatures.
Compared with the SDC–430L anode, the Ni–SDC–430L anode obtained here
shows a great decrease of the polarization resistance, especially at the lower temperatures (<700 °C) (Figs. 2.25 and 2.34b). A comparison of the Nyquist plots of
the impedance data for the SDC–430L and the Ni–SDC–430L anode measured at
650 °C is shown in Fig. 2.35. The introduction of slight Ni into the SDC–430L
anode can effectively reduce the polarization resistance from 0.230 to 0.112 X cm2.
As can be observed from Fig. 2.35, the reduction of the impedance is mainly
centered at an intermediate frequency (5–400 Hz), which normally relates to a
surface reaction [14].
The reaction pathways of the hydrogen oxidation reaction in the Ni–SDC–430L
anode is schematically shown in Fig. 2.36. Similar to the SDC–430L anode, the
430L backbone acting as the electronic pathway while the SDC particles acting as
both the ionic pathway and the reaction active sites. By adding Ni particles, TPBs
of the SDC–430L anode are further enlarged from the surfaces of the SDC particles
to the contacting areas between the SDC and Ni particles (marked in red in
2.3 Results and Discussion
45
Fig. 2.35 Impedance spectra of the symmetrical anode cells: a SDC–430L, b Ni–SDC–430L
Fig. 2.36 Schematic diagram
depicting the reaction
pathways in Ni–SDC–430L
anode
Fig. 2.36). Due to the higher catalytic activity of Ni than that of SDC, the hydrogen
oxidation reaction of the SDC–430L anode is expected to be accelerated by adding
Ni particles, which should be the reason to the fact that the Ni–SDC–430L anode
has a much lower polarization resistance than that of the SDC–430L anode
(Fig. 2.35) [14, 57].
Stability of the Ni–SDC–430L anode was also measured at 650 °C for 1200 h.
As shown in Fig. 2.37a, an increase of polarization resistance from 0.12 to 0.3 X
cm2 is observed during the initial 500 h. After that, the polarization resistance is
stabilized at around 0.3 X cm2. Impedance spectra of the anode before and after the
durability measurement are also shown in Fig. 2.37b. Consistent with the results of
the infiltrated Ni–430L anode as shown above, the increase of the polarization
resistance is mainly in the lower-frequency.
Figure 2.38 shows the SEM images of the Ni–SDC particles before and after the
durability test. After the long-term test, coarsening of the particles and loss of the
pores are shown. We surmise that the microstructure change would be the main
reason for the increase of polarization resistance during the initial 500 h. While
46
2
Fabrication and Investigation of Intermediate-Temperature …
Fig. 2.37 a Stability of the polarization resistance for the Ni–SDC–430L anode, b Nyquist plots
of the impedance data before and after the stability test. Reproduced with permission from Ref.
[27]. Copyright 2014, Elsevier
Fig. 2.38 Microstructures of the Ni–SDC particles: a Before and b After the stability test.
Reproduced with permission from Ref. [27]. Copyright 2014, Elsevier
further extending the durability measurement may not change the microstructure and
a stable polarization resistance is observed. We believe that if a pre-coarsening
process is applied at the infiltrated Ni–SDC–430L anode as that at the Ni–430L
anode, a stable polarization resistance can be obtained. Considering the low operation temperature, degradation mechanisms like metallic inter-diffusion between the
2.3 Results and Discussion
47
Fig. 2.39 SEM images of the: a MS–SOFC, b LSFSc–YSZ cathode, c Ni–SDC–430L anode.
Reproduced with permission from Ref. [15]. Copyright 2014, The Electrochemical Society
supporting alloy substrates and the Ni anode catalysts and oxidation of the porous
alloy substrates in humidified hydrogen may not be significant issues here.
Figure 2.39a shows a representative cross-sectional SEM micrograph of the
single MS–SOFC, consisting of a porous Ni–SDC infiltrated 430L anode, a dense
YSZ electrolyte and a porous LSFSc infiltrated YSZ cathode. The thickness of the
cell component is about 40, 25 and 300 µm (part of the anode is not shown in this
figure) for the cathode, electrolyte and anode, respectively. Well bonding between
the electrolyte and the adjacent layers is presented clearly. Figure 2.39b and c show
a high-magnification SEM micrograph of the LSFSc–YSZ cathode (30 wt% LSFSc
loading) and Ni–SDC–430L anode (10 wt% Ni–SDC loading), respectively. Both
the particle diameter and the pore size are 100 nm for the infiltrated LSFSc
catalysts. While for the infiltrated Ni–SDC catalysts, the average particle diameter is
20–50 nm and the mean pore size is 50 nm.
Typical cell voltages and power densities of the MS–SOFC are shown in
Fig. 2.40a as functions of current densities. The open circuit voltages decrease from
1.12 V at 600 °C to 1.07 V at 800 °C and are within 50 mV of the thermodynamically expected Nernst potentials, indicating excellent impermeability of the
YSZ electrolyte shown in Fig. 2.39a. The MPDs are 0.4, 0.68, 0.92, 1.09 and
Fig. 2.40 Electrochemical characteristics of the single cell measured at 600–800 °C: a I–P–V
characteristics, b Impedance spectra. Reproduced with permission from Ref. [15]. Copyright 2014,
The Electrochemical Society
48
2
Fabrication and Investigation of Intermediate-Temperature …
1.23 W cm−2 at 600, 650, 700, 750 and 800 °C, respectively. Compared with those
of the MS–SOFCs using infiltrated Ni or SDC as the anodes, the power densities
represented here show a significant improvement, especially at low temperatures.
Considering the 25 µm thick YSZ electrolyte applied here, the maximum power
output of 0.4 W cm−2 at 600 °C is encouraging. Such performances are even
comparable with other MS–SOFCs using SDC as the electrolytes [58–60]. Since
the oxide ion conductivity of the YSZ electrolyte used in this study is lower than
that of the SDC electrolyte, the high cell performance at low temperature of 600 °C
may be attributed to the nano-structured electrode catalysts. Figure 2.40b shows the
Nyquist plots of the impedance data obtained at open circuits for the present MS–
SOFCs. The pure ohmic resistances are 0.12, 0.15, 0.19 and 0.26 X cm2 and the
combined interfacial polarization resistances are 0.12, 0.13, 0.16 and 0.22 X cm2 at
800, 750, 700 and 650 °C, respectively. In contrast to the MS–SOFCs using
infiltrated Ni or SDC as the anode whose resistances were dominated by the
polarization part, the ohmic and the polarization resistance play a comparable role
in the cell using Ni–SDC as the anode.
Short-term stabilities of the single MS–SOFC with a cell configuration of “Ni–
SDC infiltrated 430L anode/scandia-stabilized zirconia (SSZ) electrolyte/LSFSc
infiltrated SSZ cathode” measured at 700–600 °C are shown in Fig. 2.41. SSZ was
applied here due to its higher oxide ionic conductivity than that of YSZ, especially
at lower temperatures (<700 °C). As shown in Fig. 2.41a, a rapid voltage decrease
(from 0.804 to 0.645 V) is found when measured at 700 °C with a current density
of 0.86 A cm−2. Note that reducing the operating temperature and current density
exhibit a more stable performance. As shown in Fig. 2.41b, a slight decrease of
voltage from 0.70 to 0.694 V is observed during the 357 h measurement at 650 °C
and 0.57 A cm−2. No degradation is found when further reducing the operating
temperature to 600 °C and current density to 0.4 A cm−2 (Fig. 2.41c).
I–V–P characteristics of the MS–SOFC measured after the 0, 80 and 175 h
operation at 700 °C is shown in Fig. 2.42a. A decrease of MPD from 0.72 to
0.62 W cm−2 is found during the 80 h operation. The continued operation caused a
gradual degradation, e.g., a MPD of 0.55 W cm−2 is obtained when measured after
the 175 h operation. Nyquist plots of the impedance data obtained before and after
Fig. 2.41 Stability of the MS–SOFC measured at: a 700 °C, b 650 °C, c 600 °C. Reproduced
with permission from Ref. [61]. Copyright 2015, Elsevier
2.3 Results and Discussion
49
Fig. 2.42 Electrochemical
characteristics of the MS–
SOFC before and after the
stability test measured at
700 °C: a I–P–V
characteristics, b Nyquist
plots of the impedance
spectra, c Bode plots of the
impedance spectra.
Reproduced with permission
from Ref. [61]. Copyright
2015, Elsevier
the stability test are shown in Fig. 2.42b. During the 175 h stability test, the pure
ohmic resistance (Ro) increases from 0.10 to 0.15 X cm2 while the polarization
resistance (Rp) changes from 0.24 to 0.36 X cm2. Since the Ro mainly derives from
the ohmic resistances of the electrolyte/electrodes and the interfaces between different layers, the increase of the Ro maybe caused by the oxidation of the 430L
substrate in the humidified hydrogen and/or the reduced adhesion between the 430L
support and the electrolyte [62, 63]. From Bode plots of the EIS collected at OCV
before and after the stability test (Fig. 2.42c), it is observed that the Rp change is
characterized by the increased impedance at intermediate frequencies between
100 Hz and 10 kHz. As reported, for the Ni:CGO infiltrated cermet anode, the high
frequency impedance arc was attributed to the oxide ion charge transfer resistance
between the electrolyte and the infiltrated anode (summit frequency around
500 kHz), the intermediate frequency arc (summit frequency around 300 Hz) was
ascribed to the electrochemistry of the electrode reaction, while the low frequency
50
2
Fabrication and Investigation of Intermediate-Temperature …
arc (summit frequency around 4 Hz) was shown to be related to the gas composition. Since the infiltrated particles are easy to be coarsened, we surmise that the
micrographs change of the cell electrodes which would decrease the active surface
area should be the main reason to the increase of the polarization resistance.
To verify the surmise above, SEM micrographs of the Ni–SDC infiltrated 430L
anodes and LSFSc infiltrated SSZ cathodes before and after the durability tests were
examined. As shown in Fig. 2.43a–d, coarsening of the particles and cracking of
the infiltrated coatings are clearly observed for the anodes measured after the stability tests carried out at a temperature range of 600–700 °C. In contrast, no
obvious changes in the morphologies of the LSFSc infiltrated SSZ cathodes are
observed before and after the stability tests (Fig. 2.44a–d). This is consistent with
our previous report which showed that no pronounced changes both in LSFSc
particle size and morphology were observed after the 400 h durability test measured
at 650 °C (Fig. 2.16) [31]. Based on the SEM results shown in Fig. 2.43, we can
conclude that morphological change of the infiltrated Ni–SDC coating reducing the
TPB length should be the main reason to the cell performance degradation.
Previous work showed that higher operation temperatures and higher current
densities could accelerate the coarsening of the electrodes [64–66]. That should be
the reason why the morphological change was particularly serious for the anode
tested at 700 °C and 0.86 A cm−2 (Fig. 2.43b).
Fig. 2.43 SEM images of the Ni–SDC–430L anodes: a Before the stability test and after the
stability test measured at b 700 °C, c 650 °C, d 600 °C. Reproduced with permission from Ref.
[61]. Copyright 2015, Elsevier
2.3 Results and Discussion
51
Fig. 2.44 SEM images of the LSFSc–SSZ cathodes: a Before the stability test and after the
stability test measured at b 700 °C, c 650 °C, d 600 °C. Reproduced with permission from Ref.
[61]. Copyright 2015, Elsevier
In order to identify whether inter-diffusions of Fe, Ni and Cr occurred in this
study, energy dispersive X-ray spectroscopy (EDS) spectrums of the 430L backbones before and after the stability tests were measured (Fig. 2.45). All of the
samples reflect the compositions of Fe–Cr and no Ni element is detected. It suggests
that the metal element diffusion issue may not be the problem here.
Since both the temperature and the current load are varied in Fig. 2.41, it is hard
to identify the impact of current density and temperature on degradation independently. Stabilities of the MS–SOFC measured at varied current densities and
temperatures were further studied and shown in Fig. 2.46. To evaluate the impact of
current density on degradation, we kept the temperature constant. As shown in
Fig. 2.46, the degradation rate is 9.23% (from 0.802 to 0.728 V) when measured at
700 °C and 0.86 A cm−2, while a much higher degradation rate of 15.57% (from
0.501 to 0.423 V) is found when the fuel cell operated under a higher current
density of 1.23 A cm−2. Furthermore, to evaluate the impact of temperature on
degradation, the applied current densities were kept similar (0.86 A cm−2 at 700 °C
and 0.90 A cm−2 at 650 °C). It is found that the degradation rate are 9.23% (from
0.802 to 0.728 V) and 2.61% (from 0.537 to 0.523 V) when measured at 700 °C
and 650 °C, respectively. In conclusion, both the current density and temperature
have great impact on the stability of the MS–SOFC and larger current densities and
higher temperatures would cause more significant degradation.
52
2
Fabrication and Investigation of Intermediate-Temperature …
Fig. 2.45 EDS spectra of the 430L backbone: a Before the stability test and after the stability test
measured at b 700 °C, c 650 °C, d 600 °C. Reproduced with permission from Ref. [61].
Copyright 2015, Elsevier
2.3 Results and Discussion
53
Fig. 2.45 (continued)
Fig. 2.46 Stabilities of the
MS–SOFC measured at
varied current densities and
temperatures. Reproduced
with permission from Ref.
[61]. Copyright 2015,
Elsevier
Long-term stability of the single cell operated at 650 °C under a high current
density of 0.9 A cm−2 is shown in Fig. 2.47. Voltage decrease is found in the initial
500 h while no obvious change is found during the subsequent 1000 h measurement. It is consistent with the durability test of the Ni–SDC infiltrated 430L anode,
which showed that the polarization resistance of the anode increases from 0.12 to
0.3 X cm2 during the initial 500 h while no degradation is found during the subsequent measurement at 650°C [27]. This result further confirms our conclusion
that the cell degradation is caused by the anode. As shown in Fig. 2.47, during the
1500 h measurement, a degradation rate of 1.3% kh−1 in cell voltage is found.
Another MS–SOFC with a FeCr alloy support, CGO–Ni infiltrated cermet anode,
ScYSZ electrolyte and LSCF cathode exhibited a more stable performance, i.e.,
54
2
Fabrication and Investigation of Intermediate-Temperature …
Fig. 2.47 Long-term
stability of the MS–SOFC
measured at 650 °C.
Reproduced with permission
from Ref. [61]. Copyright
2015, Elsevier
Fig. 2.48 OCV and MPD
retention of the MS–SOFCs
upon 30 thermal cycles
0.9% kh−1 at 650 °C [34]. The differences in cell durability may be caused by the
different current densities, e.g., current load of 0.9 A cm−2 was applied here while it
was only 0.25 A cm−2 in that report.
In this study, the thermal shock resistance of the MS–SOFCs was investigated
between 100 and 650 °C at a rate of 10 °C min−1. Figure 2.48 shows the OCV and
MPD retention of the MS–SOFCs upon 30 thermal cycles. Both of the OCV and
MPD show a continuous degradation during the initial 11 cycles and then tend to be
stable during the following cycles. Figure 2.49 shows the decrease of the OCV and
increase of the ohmic resistance which should be caused by the thermal stress.
Thermal shock resistance of the MS–SOFCs should be further improved by optimizing the cell structure and materials.
2.4 Conclusions
55
Fig. 2.49 Electrochemical characteristics of the MS–SOFC before and after the thermal cycles:
a I–P–V characteristics, b Impedance spectra
2.4
Conclusions
In this chapter, intermediate-temperature MS–SOFCs based on the structure of
“porous Ni/SDC/Ni–SDC infiltrated 430L anode | dense YSZ/SSZ electrolyte |
porous LSFSc infiltrated YSZ/SSZ cathode” were developed. The
structure-performance relationship of the electrodes, reaction kinetics of the electrodes and degradation mechanism of the single cells were also investigated. The
specific conclusions are listed as follows:
1. LSFSc cathode and Ni-based anode materials were applied by the infiltration
method to reduce the polarization resistances of the cathode and anode,
respectively. When measured at 650 °C, polarization resistances of the infiltrated LSFSc–YSZ cathode, Ni–430L anode, SDC–430L anode and Ni–SDC–
430L anode were 0.160, 2.2, 0.233 and 0.112 X cm2, respectively.
2. Coarsening of the Ni particles and elemental inter-diffusion between Ni and
430L would cause the degradation of the Ni–430L anode.
3. Kinetic studies indicated that the low frequency peak of the impedance spectra
of the SDC–430L anode was associated with the extraction of surface lattice
oxygen of SDC while the high frequency peak was attributed to the transport of
oxide-ions within the SDC coating.
4. MPD of the MS–SOFC with Ni–SDC–430L anode were 1.23, 0.92 and
0.40 W cm−2 when measured at 800, 700 and 600 °C, respectively.
A degradation rate of 1.3% kh−1 was shown when measured at 650 °C for
1500 h. The degradation rate was faster at higher operation temperatures and
larger current densities and the degradation was mainly caused by the morphological change of the anode.
56
2
Fabrication and Investigation of Intermediate-Temperature …
References
1. Tucker MC (2010) Progress in metal-supported solid oxide fuel cells: a review. J Power
Sources 195:4570–4582
2. Tucker MC, Lau GY, Jacobson CP, DeJonghe LC, Visco SJ (2008) Stability and robustness
of metal-supported SOFCs. J Power Sources 175:447–451
3. Leah RT, Brandon NP, Aguiar P (2005) Modelling of cells, stacks and systems based around
metal-supported planar IT-SOFC cells with CGO electrolytes operating at 500–600 °C.
J Power Sources 145:336–352
4. Brandner M, Bram M, Froitzheim J, Buchkremer HP, Stoever D (2008) Electrically
conductive diffusion barrier layers for metal-supported SOFC. Solid State Ionics 179:1501–
1504
5. Franco T, Schibinger K, Ilhan Z, Schiller G, Venskutonis A (2007) Ceramic diffusion barrier
layers for metal supported SOFCs. In: Eguchi K, Singhai SC, Yokokawa H, Mizusaki H
(eds) Solid oxide fuel cells, vol 10, pp 771–780
6. Hwang C, Tsai C-H, Lo C-H, Sun C-H (2008) Plasma sprayed metal supported YSZ/Ni–
LSGM–LSCF ITSOFC with nanostructured anode. J Power Sources 180:132–142
7. Ishihara T, Yan J, Enoki M, Okada S, Matsumoto H (2008) Ni-Fe alloy-supported
intermediate temperature SOFCs using LaGaO3 electrolyte film for quick startup. J Fuel Cell
Sci Technol 5:031205
8. Nedelec R, Neagu R, Uhlenbruck S, Maric R, Sebold D, Buchkremer HP et al (2011) Gas
phase deposition of diffusion barriers for metal substrates in solid oxide fuel cells. Surf Coat
Technol 205:3999–4004
9. Ju Y-W, Hong J-E, Hyodo J, Inagaki T, Ida S, Ishihara T (2012) New buffer layer material La
(Pr)CrO3 for intermediate temperature solid oxide fuel cell using LaGaO3-based electrolyte
film. J Mater Res 27:1906–1914
10. Hui R, Berghaus JO, Deces-Petit C, Qu W, Yick S, Legoux J-G et al (2009) High
performance metal-supported solid oxide fuel cells fabricated by thermal spray. J Power
Sources 191:371–376
11. Yoo Y, Wang Y, Deng X, Singh D, Legoux J-G (2012) Metal supported tubular solid oxide
fuel cells fabricated by suspension plasma spray and suspension high velocity oxy-fuel spray.
J Power Sources 215:307–311
12. Vohs JM, Gorte RJ (2009) High-performance SOFC cathodes prepared by infiltration. Adv
Mater 21:943–956
13. Blennow P, Hjelm J, Klemenso T, Ramousse S, Kromp A, Leonide A et al (2011)
Manufacturing and characterization of metal-supported solid oxide fuel cells. J Power Sources
196:7117–7125
14. Blennow P, Hjelm J, Klemenso T, Persson AH, Ramousse S, Mogensen M (2011) Planar
metal-supported SOFC with novel cermet anode. Fuel Cells 11:661–668
15. Zhou Y, Ye X, Li J, Zhan Z, Wang S (2014) Metal-supported solid oxide fuel cells with a
simple structure. J Electrochem Soc 161:F332–F336
16. Chen J, Liang FL, Liu LN, Jiang SP, Chi B, Pu J et al (2008) Nano-structured (La, Sr)(Co, Fe)
O3 + YSZ composite cathodes for intermediate temperature solid oxide fuel cells. J Power
Sources 183:586–589
17. Huang Y, Vohs JM, Gorte RJ (2004) Fabrication of Sr-doped LaFeO3 YSZ composite
cathodes. J Electrochem Soc 151:A646–A651
18. Zhang Y, Sun Q, Xia C, Ni M (2013) Geometric properties of nanostructured solid oxide fuel
cell electrodes. J Electrochem Soc 160:F278–F289
19. Sholklapper TZ, Kurokawa H, Jacobson C, Visco S, De Jonghe L (2007) Nanostructured
solid oxide fuel cell electrodes. Nano Lett 7:2136–2141
20. Siebert E, Hammouche A, Kleitz M (1995) Impedance spectroscopy analysis of
La1−xSrxMnO3-yttria-stabilized zirconia electrode kinetics. Electrochim Acta 40:1741–1753
References
57
21. Tsuneyoshi K, Mori K, Sawata A, Mizusaki J, Tagawa H (1989) Kinetic studies on the
reaction at the La0.6Ca0.4MnO3/YSZ interface, as an SOFC air electrode. Solid State Ionics
35:263–268
22. Takeda Y, Kanno R, Noda M, Tomida Y, Yamamoto O (1987) Cathodic polarization
phenomena of perovskite oxide electrodes with stabilized zirconia. J Electrochem Soc
134:2656–2661
23. Escudero MJ, Aguadero A, Alonso JA, Daza L (2007) A kinetic study of oxygen reduction
reaction on La2NiO4 cathodes by means of impedance spectroscopy. J Electroanal Chem
611:107–116
24. Wang Y, Zhang L, Chen F, Xia C (2012) Effects of doped ceria conductivity on the
performance of La0.6Sr0.4Co0.2Fe0.8O3−d cathode for solid oxide fuel cell. Int J Hydrogen
Energy 37:8582–8591
25. Kim JD, Kim GD, Moon JW, Park YI, Lee WH, Kobayashi K et al (2001) Characterization of
LSM-YSZ composite electrode by ac impedance spectroscopy. Solid State Ionics 143:379–
389
26. Li Q, Xia T, Sun L, Zhao H, Huo L (2014) Electrochemical performance of novel cobalt-free
perovskite SrFe0.7Cu0.3O3−d cathode for intermediate temperature solid oxide fuel cells.
Electrochim Acta 150:151–156
27. Zhou Y, Yuan C, Chen T, Meng X, Ye X, Li J et al (2014) Evaluation of Ni and Ni–
Ce0.8Sm0.2O2−d (SDC) impregnated 430L anodes for metal-supported solid oxide fuel cells.
J Power Sources 267:117–122
28. Liu X, Meng X, Han D, Wu H, Zeng F, Zhan Z (2013) Impregnated nickel anodes for
reduced-temperature solid oxide fuel cells based on thin electrolytes of doped LaGaO3.
J Power Sources 222:92–96
29. Chen M, Kim BH, Xu Q, Ahn BG, Huang DP (2010) Effect of Ni content on the
microstructure and electrochemical properties of Ni–SDC anodes for IT-SOFC. Solid State
Ionics 181:1119–1124
30. Jiang S, Badwal S (1999) An electrode kinetics study of H2 oxidation on Ni/Y2O3-ZrO2
cermet electrode of the solid oxide fuel cell. Solid State Ionics 123:209–224
31. Zhou Y, Xin X, Li J, Ye X, Xia C, Wang S et al (2014) Performance and degradation of
metal-supported solid oxide fuel cells with impregnated electrodes. Int J Hydrogen Energy
39:2279–2285
32. Ansar A, Szabo P, Arnold J, Ilhan Z, Soysal D, Costa R et al (2011) Metal supported solid
oxide fuel cells and stacks for auxiliary power units—progress, challenges and lessons
learned. Solid Oxide Fuel Cells 12 (Sofc Xii). 35:147–55
33. Soysal D, Ansar A, Ilhan Z, Costa R (2011) Nanostructured composite cathodes by
suspension plasma spraying for SOFC applications. In: Singhal SC, Eguchi K (eds) Solid
oxide fuel cells, vol 12, pp 2233–2241
34. Klemenso T, Nielsen J, Blennow P, Persson AH, Stegk T, Christensen BH et al (2011) High
performance metal-supported solid oxide fuel cells with Gd-doped ceria barrier layers.
J Power Sources 196:9459–9466
35. Simner SP, Shelton JP, Anderson MD, Stevenson JW (2003) Interaction between La
(Sr) FeO3 SOFC cathode and YSZ electrolyte. Solid State Ionics 161:11–18
36. Liu Y, Chen K, Zhao L, Chi B, Pu J, Jiang SP et al (2014) Performance stability and
degradation mechanism of La0.6Sr0.4Co0.2Fe0.8O3−d cathodes under solid oxide fuel cells
operation conditions. Int J Hydrogen Energy 39:15868–15876
37. Hathiramani D, Vaßen R, Mertens J, Sebold D, Haanappel VAC, Stöver D (2008)
Degradation mechanism of metal supported atmospheric plasma sprayed solid oxide fuel
cells. In: Advances in solid oxide fuel cells II: ceramic engineering and science proceedings.
Wiley, New York, pp 55–65
38. Franco T, HoshiarDin Z, Szabo P, Lang M, Schiller G (2007) Plasma sprayed diffusion barrier
layers based on doped perovskite-type LaCrO3 at substrate-anode interface in solid oxide fuel
cells. J Fuel Cell Sci Technol 4:406–412
58
2
Fabrication and Investigation of Intermediate-Temperature …
39. Choi JH, Lee T, Choi M, Yoo Y-S, Baek S-W, Bae J (2010) Long-term performance of
anode-supported SOFC integrated with metal interconnect by joining process. Int J Hydrogen
Energy 35:4285–4291
40. Shaigan N, Qu W, Ivey DG, Chen W (2010) A review of recent progress in coatings, surface
modifications and alloy developments for solid oxide fuel cell ferritic stainless steel
interconnects. J Power Sources 195:1529–1542
41. Fergus JW (2005) Metallic interconnects for solid oxide fuel cells. Mater Sci Eng A 397:
271–283
42. Molin S, Kusz B, Gazda M, Jasinski P (2008) Evaluation of porous 430L stainless steel for
SOFC operation at intermediate temperatures. J Power Sources 181:31–37
43. Lee JG, Park JH, Shul YG (2014) Tailoring gadolinium-doped ceria-based solid oxide fuel
cells to achieve 2 W cm−2 at 550 °C. Nat Commun 5:4045
44. Ding C, Hashida T (2010) High performance anode-supported solid oxide fuel cell based on
thin-film electrolyte and nanostructured cathode. Energy Environ Sci 3:1729–1731
45. Chen Y, Liu Q, Yang Z, Chen F, Han M (2012) High performance low temperature solid
oxide fuel cells with novel electrode architecture. RSC Adv 2:12118–12121
46. Xia C, Liu M (2001) Low-temperature SOFCs based on Gd0.1Ce0.9O1.95 fabricated by dry
pressing. Solid State Ionics 144:249–255
47. Désaunay T, Bonura G, Chiodo V, Freni S, Couzinié JP, Bourgon J et al (2013)
Surface-dependent oxidation of H2 on CeO2 surfaces. J Catal 297:193–201
48. Chen B, Ma Y, Ding L, Xu L, Wu Z, Yuan Q et al (2013) Reactivity of hydroxyls and water
on a CeO2 (111) thin film surface: the role of oxygen vacancy. J Phys Chem C 117:
5800–5810
49. Zhou Y, Luo T, Du X, Wang J, Yang W, Sun C et al (2014) High activity of nanoporous—
Sm0.2Ce0.8O2‐d@ 430L composites for hydrogen electro—oxidation in solid oxide fuel cells.
Adv Energy Mater 4:1400883
50. Hagen A, Liu YL, Barfod R, Hendriksen PV (2007) Anode supported solid oxide fuel cells—
deconvolution of degradation into cathode and anode contributions. ECS Trans 7:301–309
51. Bastidas DM, Tao S, Irvine JTS (2006) A symmetrical solid oxide fuel cell demonstrating
redox stable perovskite electrodes. J Mater Chem 16:1603–1605
52. Liu Q, Dong X, Xiao G, Zhao F, Chen F (2010) A novel electrode material for symmetrical
SOFCs. Adv Mater 22:5478–5482
53. Gross MD, Vohs JM, Gorte RJ (2007) A strategy for achieving high performance with SOFC
ceramic anodes. Electrochem Solid-State Lett 10:B65–B69
54. Alam MK, Ahmed F, Miura R, Suzuki A, Tsuboi H, Hatakeyama N et al (2011) Surface
reduction processes of cerium oxide surfaces by H2 using ultra accelerated quantum chemical
molecular dynamic study. Catal Today 164:9–15
55. Zhang C, Yu Y, Grass ME, Dejoie C, Ding W, Gaskell K et al (2013) Mechanistic studies of
water electrolysis and hydrogen electro-oxidation on high temperature ceria-based solid oxide
electrochemical cells. J Am Chem Soc 135:11572–11579
56. Souza ECC, Chueh WC, Jung W, Muccillo ENS, Haile SM (2012) Ionic and electronic
conductivity of nanostructured, samaria-doped ceria. J Electrochem Soc 159:K127–K135
57. Nielsen J, Klemenso T, Blennow P (2012) Detailed impedance characterization of a well
performing and durable Ni:CGO infiltrated cermet anode for metal-supported solid oxide fuel
cells. J Power Sources 219:305–316
58. Xie Y, Neagu R, Hsu C-S, Zhang X, Deces-Petit C, Qu W et al (2010) Thin film solid oxide
fuel cells deposited by spray pyrolysis. J Fuel Cell Sci Technol 7:021007
59. Oberste Berghaus J, Legoux JG, Moreau C, Hui R, Deces-Petit C, Qu W et al (2008)
Suspension HVOF spraying of reduced temperature solid oxide fuel cell electrolytes. J Therm
Spray Technol 17:700–707
60. Oishi N, Yoo Y, Davidson I (2007) Fabrication of a metal-supported ceria cell by wet powder
spray and its characterization at intermediate temperatures. ECS Trans 7:781–786
References
59
61. Zhou Y, Chen T, Li J, Yuan C, Xin X, Chen G et al (2015) Long-term stability of
metal-supported solid oxide fuel cells employing infiltrated electrodes. J Power Sources
295:67–73
62. Knibbe R, Wang H-J, Blennow P, Thydén K, Persson ÅH, Mikkelsen L et al (2013)
Oxidation in ceria infiltrated metal supported SOFCs—a TEM investigation. J Power Sources
228:75–82
63. Sarasketa-Zabala E, Otaegi L, Rodriguez-Martinez LM, Alvarez MA, Burgos N, Castro F
et al (2012) High temperature stability of porous metal substrates under highly humidified
hydrogen conditions for metal supported Solid Oxide Fuel Cells. Solid State Ionics 222:16–22
64. Meng X, Han D, Wu H, Li J, Zhan Z (2014) Characterization of SrFe0.75Mo0.25O3− d–
La0.9Sr0.1Ga0.8Mg0.2O3−d composite cathodes prepared by infiltration. J Power Sources
246:906–911
65. Komatsu T, Yoshida Y, Watanabe K, Chiba R, Taguchi H, Orui H et al (2010) Degradation
behavior of anode-supported solid oxide fuel cell using LNF cathode as function of current
load. J Power Sources 195:5601–5605
66. Hatae T, Matsuzaki Y, Yamashita S, Yamazaki Y (2009) Current density dependence of
changes in the microstructure of SOFC anodes during electrochemical oxidation. Solid State
Ionics 180:1305–1310
Chapter 3
Fabrication and Investigation
of Low-Temperature MS–SOFCs
3.1
Introduction
In the previous chapter, intermediate-temperature MS–SOFCs based on the infiltrated LSFSc cathodes and Ni based anodes have been successfully developed and
promising output power densities and long-term stabilities have been achieved at the
intermediate-temperatures (600–800 °C). However, there are some major challenges
remained to be addressed due to the high operating temperatures, e.g., oxidation of
the porous stainless steels, metallic inter-diffusion between the substrate and the Ni
based anode, and coarsening of the infiltrated nano particles. Reducing the operating
temperatures (<600 °C) of MS–SOFCs is expected to be an effective way to solve
these issues. During the long-term development of SOFCs, developing new materials, e.g., strontium and magnesium doped lanthanum gallate (LSGM), doped bismuth oxides, Ba0.5Sr0.5Co0.8Fe0.2O3−d (BSCF), and optimizing cell structures, e.g.,
reducing the thickness of electrolytes and designing micro-nano structured electrodes
are the two key routes to improve cell performances and reduce the operating temperatures [1–7]. As shown in the pervious chapter, infiltrated Ni–SDC–430L anode
exhibited a low activation energy of 0.31 eV at 600–800 ºC, while that of the
LSFSc–YSZ was as high as 1.10 eV at 650–800 ºC. Thus, optimizing the cathode
materials to reduce the polarization resistance of the cathode is key to obtain a
low-temperature operated MS–SOFC. Compared with La1–xSrxFeO3−d (LSF),
Co-based cathode materials show a higher electrochemical performance especially at
low temperatures due to the high electronic and ionic conductivities, and enhanced
surface exchange rates of oxygen [8]. In this chapter, electrochemical performances
and long-term stabilities of the infiltrated La0.8Sr0.2CoO3−d (LSC), La0.58Sr0.4Co0.2
Fe0.8O3−d (LSCF) and SmBa0.5Sr0.5Co2.0O5+d (SBSC) cathodes were investigated.
In addition, MS–SOFCs based on the infiltrated SBSC-scandia stabilized zirconia
(SSZ) cathode have been developed and the performances were studied.
© Springer Nature Singapore Pte Ltd. 2018
Y. Zhou, Study on Fabrication and Performance of Metal-Supported Solid Oxide
Fuel Cells, Springer Theses, https://doi.org/10.1007/978-981-10-6617-7_3
61
62
3.2
3.2.1
3 Fabrication and Investigation of Low-Temperature MS–SOFCs
Experimental Section
Fabrication of Symmetric and Single Cells
For single cell preparation, commercial 430L stainless steel powder (−400 mesh,
Jing-yuan Powder Material Co., Ltd, China) and SSZ powder (Tosoh Corporation,
Japan) were used as starting materials. The slurry for tape casting was ethanol based
which contained pore-forming agent, dispersing agent, binder, plasticizer and other
organic additives, in addition to powders. The simple tri-layer structure of porous
430L | SSZ electrolyte | porous SSZ backbone was produced by laminating tape
cast green tapes and subsequent co-firing at 1300 ºC for 4 h in a reducing atmosphere (5% H2/95% N2). Symmetric cathode cells based upon the configuration of
“porous SSZ/Ce0.8Sm0.2O2−d (SDC) | SSZ electrolyte | porous SSZ/SDC” were
prepared similarly.
For symmetric cathode cells, precursor solutions of LSC, LSCF and SBSC were
infiltrated into the porous SSZ scaffold and that of LSCF was infiltrated into the
porous Ce0.8Sm0.2O2−d (SDC) scaffold, followed by calcining at 700 °C for 2 h.
The infiltration-calcination process was repeated to achieve desired loadings of
the cathode materials. A single infiltration/calcination cycle yielded a loading of
5 wt% and the ultimate loading was controlled at 30 wt%. For the single cells, the
anode precursor solution with Ni(NO3)2, Sm(NO3)3 and Ce(NO3)3 in stoichiometric
ratios (the mass ratio of SDC:Ni = 8:2) was infiltrated into the anode side while the
cathode solution was infiltrated into the cathode side. After drying, heat treatment
was conducted in air for 1 h at 350 ºC. For the final step to form the phases, the
ultimate sintering process of the infiltrated electrodes was carried out at 700 ºC for
2 h during the cell testing process. For the anode, a single infiltration/calcination
cycle yielded a loading of 2–3 wt% and the ultimate loading was 10 wt%.
3.2.2
Material Characterizations
Phase compositions of the infiltrated cathodes/cathode powders were identified at
room temperature using a Rigaku XRD diffractometer with monochromatic CuKa
radiation. The microstructure of the cells was examined using scanning electron
microscopy (SEM) in Hitachi S–4800–II and SU–8220 microscopes.
3.2.3
Electrochemical Measurements
For electrochemical measurements, silver paste (DAD–87, Shanghai Research
Institute of Synthetic Resins) was applied onto the electrode surfaces as the current
collector and silver wires were attached as the voltage and current leads.
3.2 Experimental Section
63
Current-voltage curves were obtained using an IM6 Electrochemical Workstation
(ZAHNER, Germany) at 600–800 ºC with cathodes exposed to air and anodes to
humidified (3% H2O) hydrogen both at 100 sccm. Electrochemical impedance
spectra (EIS) were collected at open circuits with the signal amplitude of 20 mV
and in the frequency range of 200 kHz to 0.1 Hz over a temperature range of 550–
700 °C. Impedance measurements were also performed in air on symmetric cathode
cells. The ohmic resistance (Ro) was obtained from the value of the high frequency
intercept. Area specific polarization resistance (Rp) was determined by the difference of the low and high frequency intercepts of the impedance spectra with the real
axis. Active areas of the symmetric cathode cell and the single cell were 0.35 cm2.
3.3
3.3.1
Results and Discussion
Investigation of Infiltrated LSC/LSCF/SBSC–SSZ
Cathodes
Figure 3.1a shows a representative cross-sectional SEM micrograph of the symmetric cathode cell before infiltration, consisting of a porous SSZ backbone, a dense
SSZ electrolyte and a porous SSZ backbone. The electrolyte is fully dense with a
thickness of 250 lm and is well bonded with the adjacent porous SSZ layers. The
thickness of the porous SSZ layer is about 30 lm. Figure 3.1b shows a
higher-magnification SEM micrograph of the porous and dense SSZ layers and a
good connection between the two layers is obviously exhibited.
XRD patterns of the LSC, LSCF and SBSC powder calcined at 600, 700 and
800 °C are shown in Fig. 3.2a–c, respectively. The dominant phases, perovskite
LSC, LSCF and SBSC are formed at the temperature above 700 °C. Small amount
of impurities like SrCO3 in LSC and LSCF, and BaCoO3 in SBSC can still be found
Fig. 3.1 SEM images of the symmetrical cathode backbone. Reproduced with permission
from Ref. [26]. Copyright 2014, Elsevier
64
3 Fabrication and Investigation of Low-Temperature MS–SOFCs
Fig. 3.2 XRD patterns of the powder calcined at 600–800 °C: a LSC, b LSCF, c SBSC.
Reproduced with permission from Ref. [27]. Copyright 2015, Elsevier
even after increasing the calcining temperature to 800 °C. In order to avoid
coarsening of the infiltrated nano particles and possible solid reactions between the
cathode and the electrolyte materials at a high temperature, the calcining temperature of infiltrated LSC–SSZ, LSCF–SSZ and SBSC–SSZ cathodes was selected at
700 °C in this study.
Figure 3.3 shows the impedance spectra of the infiltrated cathodes measured in
air at 700 °C. As shown in Fig. 3.3a, the Rp values of SBSC–SSZ, LSC–SSZ and
LSCF–SSZ are 0.054, 0.084 and 0.140 X cm2, respectively. From the Bode plots of
Fig. 3.3 Impedance spectra of the infiltrated cathodes: a Nyquist plots, b Bode plots. Reproduced
with permission from Ref. [27]. Copyright 2015, Elsevier
3.3 Results and Discussion
65
the impedance spectra shown in Fig. 3.3b, the differences in EIS of different
infiltrated cathodes mainly lie in high and medium frequencies. Since the high
frequency is usually attributed to the oxygen ion transfer process from the TPB to
the electrolyte and the intermediate frequency peak is associated with the surface
kinetics, the differences of the EIS plots can be explained by the different charge
transfer and surface processes [9, 10]. As reported, electronic and ionic conductivity
of La0.6Sr0.4Co0.2Fe0.8O3−d at 800 °C was 302 S cm−1 and 8 10−3 S cm−1,
respectively. For La0.6Sr0.4CoO3−d, the electronic and ionic conductivity (800 °C)
could be as high as 1.58 103 S cm−1 and 0.22 S cm−1, respectively [8]. Since the
higher conductivity would facilitate the charge transfer process at the interface
of cathode and electrolyte, there is no wonder that the polarization resistance of
LSC–SSZ is lower than that of LSCF–SSZ. Fukunaga et al. reported that the
oxygen adsorption and desorption processes at the surface of Sm0.5Sr0.5CoO3 was
one order of magnitude larger than the corresponding values calculated for
La0.6Sr0.4CoO3, which may explain why SBSC–SSZ cathode exhibited the lowest
Rp result as shown in Fig. 3.3a [11].
The polarization resistances measured at temperatures from 550 to 700 °C for
the infiltrated cathodes are summarized in Fig. 3.4. It shows that the RP values of
the cathodes decrease in the following order: LSCF–SSZ > LSC–SSZ > SBSC–
SSZ at all measured temperatures. The activation energies of the resistances of the
LSC–SSZ, LSCF–SSZ and SBSC–SSZ cathodes, calculated from the Arrhenius
plots of the fitted line, are 1.20, 1.06 and 1.05 eV, respectively.
Long-term stabilities of the infiltrated cathodes were measured and the evolution
of Ro and Rp values with testing time are shown in Fig. 3.5. LSC–SSZ and LSCF–
SSZ samples were tested at 620 °C for 1400 h and the SBSC–SSZ sample was
tested for 820 h. The differences in the initial ohmic losses measured for LSC–SSZ,
LSCF–SSZ and SBSC–SSZ are ascribed to the differences in the electrolyte
thickness and the electronic conductivity of infiltrated materials. The polarization
resistances of LSC–SSZ, LSCF–SSZ and SBSC–SSZ cathodes gradually increase
from 0.333 to 1.171 X cm2, 0.347 to 0.609 X cm2, and 0.186 to 0.328 X cm2, at an
Fig. 3.4 Activation energies
of the polarization resistances
of the infiltrated cathodes.
Reproduced with permission
from Ref. [27]. Copyright
2015, Elsevier
66
3 Fabrication and Investigation of Low-Temperature MS–SOFCs
Fig. 3.5 Stabilities of the infiltrated cathodes: a Polarization and b Ohmic resistances.
Reproduced with permission from Ref. [27]. Copyright 2015, Elsevier
average degradation rate of 179, 54 and 93% kh−1, respectively. While, the ohmic
resistances of LSC–SSZ, LSCF–SSZ and SBSC–SSZ cathodes increase from 1.561
to 1.792 X cm2, 2.121 to 2.384 X cm2 and 1.176 to 1.570 X cm2, at an average
degradation rate of 11, 9 and 41% kh−1, respectively. The stability of Rp decreases
in the following order: LSCF–SSZ > SBSC–SSZ > LSC–SSZ, with Ro decreasing
as LSCF–SSZ > LSC–SSZ > SBSC–SSZ. The increase in Rp is much higher than
that in Ro, indicating that polarization resistances dominate the degradation. In
order to determine the degradation mechanism, morphology evolutions and solidstate reactions of the infiltrated cathodes were investigated.
Figure 3.6 shows the impedance spectra of the infiltrated cathodes measured
with respect to elapsing time. It is seen that the impedance increase gradually and
Fig. 3.6 Impedance spectra of the infiltrated cathodes measured before and after the stability test:
a LSC–SSZ, b LSCF–SSZ, c SBSC–SSZ. Reproduced with permission from Ref. [27]. Copyright
2015, Elsevier
3.3 Results and Discussion
67
Fig. 3.7 SEM images of the cathodes before the stability test: a1 LSC–SSZ, b1 LSCF–SSZ, c1
SBSC–SSZ infiltrated SSZ; SEM images of the cathodes after the stability test: a2 LSC–SSZ, b2
LSCF–SSZ, c2 SBSC–SSZ. Reproduced with permission from Ref. [27]. Copyright 2015,
Elsevier
the increase occurs mainly at the intermediate frequency, which is usually associated with the surface kinetics. Thus, the increase of the frequency peak may be
explained by the loss of surface-active areas for oxygen reduction.
Figure 3.7 shows the SEM micrographs of the LSC–SSZ, LSCF–SSZ and
SBSC–SSZ electrodes before and after the stability test. Although the calcination of
the LSC, LSCF and SBSC infiltrated SSZ scaffold was conducted at the same
temperature, the morphology of infiltrated particles are quite different from each
other, as shown in Fig. 3.7a1, b1 and c1. The micrographs show that LSC and
SBSC particles are evenly distributed on the surface of the SSZ scaffolds with the
particle size of 50–100 nm, whereas LSCF particles tend to be aggregations, sizing
from 50 to 500 nm. After the stability test, shown in Fig. 3.7a2, b2 and c2, the LSC
and LSCF particles are no longer easily distinguishable, appearing to form a dense
polycrystalline layer over the SSZ scaffolds with reduced porosities. The SBSC
particles grow dramatically after 820 h, forming rod-like crystallites with the length
68
3 Fabrication and Investigation of Low-Temperature MS–SOFCs
of 0.1–0.8 lm, agglomerated on the SSZ scaffold. The morphological changes of
LSC, LSCF and SBSC decrease the surface areas of the infiltrated phases, which
decrease the TPB lengths and porosities, hinder the gas transport process, and
thereby increase the Rp values. These results are consistent with the EIS changes
shown in Fig. 3.6. Furthermore, the agglomeration of infiltrated particles leads to
discrete distribution network on the porous scaffold, disconnects the current path
and thus increases the Ro values.
To detect the potential reactions between the infiltrated cathode materials and the
SSZ scaffold, the possible solid reaction between the cathode powder and the SSZ
powder was evaluated. The powder of LSC, LSCF and SBSC was synthesized by
calcining at 700 °C for 2 h in air. XRD results shown in Fig. 3.8a1, b1 and c1 reveal
that no obvious solid reaction occurred for the mixture of LSC + SSZ, LSCF + SSZ
and SBSC + SSZ after the 930 h tests at 620 °C, except for the impurities caused by
the low calcining temperature of the starting cathode materials as shown in Fig. 3.2.
Fig. 3.8 XRD patterns of the mixture powder calcined at 620 °C for 930 h: a1 LSC–SSZ, b1
LSCF–SSZ, c1 SBSC–SSZ; calcined at 800 °C for 100 h: a2 LSC–SSZ, b2 LSCF–SSZ, c2
SBSC–SSZ. Reproduced with permission from Ref. [27]. Copyright 2015, Elsevier
3.3 Results and Discussion
69
Therefore, solid reaction is not the reason for the deactivation of the cell performance
measured at 620 °C. For comparison, the solid reaction becomes much more
obvious by increasing the calcining temperature to 800 °C. As shown in Fig. 3.8a2,
b2 and c2, after the 100 h heat treatment at 800 °C, secondary phases like SrZrO3
and La2Zr2O7 are shown in the mixture of LSC + SSZ powder, while Co3O4 and
SrZrO3 are found in LSCF + SSZ and SmZrO3, SrZO3 and Co3O4 in
SBSCO + SSZ mixtures. Based upon these XRD results, we can conclude that the
increase in Rp and Ro are contributed to the coarsening of the infiltrated particles, not
to the formation of insulating phases. Previous studies on the degradation mechanism of the infiltrated cathodes also suggested that coarsening of infiltrated
nanoparticles was the main reason to the increased resistance [12–14].
In order to further confirm the conclusion above, LSCF infiltrated SDC cathode
was fabricated and the long-term stability was measured. Since SDC has a good
chemical compatibility with the perovskite cathode materials, solid reaction would
not be the problem in the LSCF–SDC cathode [15, 16]. Figure 3.9 shows
the impedance values of LSCF infiltrated SDC symmetric cells measured in air at
650 °C for 820 h. The Rp and Ro results gradually increase from 0.346 to 1.142 X
cm2, and 4.655 to 6.084 X cm2, at an average increasing rate of 280.6 and 37.4%
kh−1, respectively. Figure 3.10 shows the morphology of the infiltrated LSCF
Fig. 3.9 Stability of the LSCF–SDC cathode: a Polarization and b Ohmic resistances.
Reproduced with permission from Ref. [27]. Copyright 2015, Elsevier
Fig. 3.10 SEM images of the LSCF particles: a Before and b after the stability test. Reproduced
with permission from Ref. [27]. Copyright 2015, Elsevier
70
3 Fabrication and Investigation of Low-Temperature MS–SOFCs
particles before and after the stability test. After the 820 h test, well-connected
LSCF particles tend to coarsen and form a dense layer. The stability test of the
LSCF–SDC cathode further confirms that the degradation of the cathode is mainly
caused by the morphological change of the infiltrated particles.
In conclusion, coarsening of the infiltrated particles is the main reason to the
degradation of infiltrated cathodes. Reducing operating temperatures, adding sintering inhibition agents and optimizing electrode structures are reported to improve
the stability of the infiltrated electrodes [14, 17–19].
3.3.2
Investigation of Infiltrated SBSC–SSZ Cathodes
and the MS–SOFCs
As shown above, the SBSC–SSZ cathode showed the lowest polarization resistance. In this section, electrochemical performances of the SBSC–SSZ cathode
were further evaluated and the MS–SOFCs based on such cathode were developed
and investigated.
Figure 3.11 shows a high-magnification SEM micrograph of the infiltrated
SBSC particles calcinated at 700, 750 and 800 ºC, respectively. Nano structured
Fig. 3.11 SEM images of the SBSC particles calcined at a 700 °C, b 750 °C, c 800 °C.
Reproduced with permission from Ref. [26]. Copyright 2014, Elsevier
3.3 Results and Discussion
71
SBSC coatings are shown in all the temperature range and the infiltrated particles
tend to be larger and denser as the calcination temperature increased. Similar
behavior has also been observed in other infiltrated electrodes [20, 21]. As shown in
Fig. 3.11, the average particle sizes of the SBSC catalysts calcinated at 700, 750
and 800 ºC are estimated to be 50, 100 and 200 nm, respectively.
Nyquist plots of the impedance data obtained at 700 ºC for the symmetric
cathode cells calcinated at 700–800 ºC are shown in Fig. 3.12a. The cathode
polarization resistances are 0.054, 0.079 and 0.117 X cm2 when calcinated at 700,
750 and 800 °C, respectively. As shown in Fig. 3.12a, all impedance arcs consist of
a large higher-frequency arc and a small lower-frequency arc. The higher-frequency
arc increases much more remarkably than the lower-frequency arc as the calcining
temperature increased. This phenomenon is more obvious in the bode representation of the EIS (Fig. 3.12b) which shows an increase in the frequency peak from
the intermediate frequency at about 10 Hz to the high frequency at 105 Hz. Since
the intermediate frequency peak is usually associated with the surface kinetics
and the high frequency peak is attributed to the oxygen ion transfer process from
the TPBs to the electrolyte, the increase of the frequency peak can be explained
by the loss of surface active areas for oxygen reduction and TPBs for charge
transfer as the calcining temperature increased [9, 10]. This explanation is further
supported by the SEM micrograph of the infiltrated SBSC particles shown in
Fig. 3.11.
Polarization resistances measured at temperatures from 550 to 700 °C for the
infiltrated SBSC–SSZ cathodes calcinated at various temperatures are summarized
in Fig. 3.13 and smaller RP values at all measurement temperatures for the cathode
calcinated at lower temperatures are exhibited. The activation energies for oxygen
reduction reaction of the SBSC–SSZ cathodes calcinated at 700, 750 and 800 °C
are 0.94, 0.87 and 0.98 eV, respectively.
The effect of thermal cycling on the polarization resistance of the infiltrated
SBSC–SSZ cathode (calcinated at 700 °C) was investigated. The thermal cycling
experiment was carried out between 100 and 600 ºC, at a rate of 10 ºC min−1.
Fig. 3.12 Impedance spectra of the polarization resistances of SBSC–SSZ cathodes calcined at
700, 750 and 800 °C: a Nyquist plots, b Bode plots. Reproduced with permission from Ref. [26].
Copyright 2014, Elsevier
72
3 Fabrication and Investigation of Low-Temperature MS–SOFCs
Fig. 3.13 Activation energies of the resistances of the cathodes calcined at 700–800 ºC.
Reproduced with permission from Ref. [26]. Copyright 2014, Elsevier
Despite the initial increase and some variation, the polarization resistance is stabilized at around 0.17 X cm2 at 600 ºC and no degradation is found after the 35
cycles (Fig. 3.14a). This result is consistent with the EIS plots of the cathode
polarization resistance measured before and after the thermal cycling test as shown
in Fig. 3.14b. Note that the coefficient of thermal expansion (CTE) of SBSC can be
as high as 21.9 10−6 K−1 from room temperature to 700 °C [22], much higher
than the traditional electrolyte materials, e.g., 12.2 10−6 K−1 at 50–800 °C for
SDC, and 10.4 10−6 K−1 at 30–750 °C for SSZ [23, 24]. The good thermal
shock resistance of the SBSC–SSZ cathode shown here should be attributed to the
improved CTE match obtained by the infiltrating method [25].
Electrochemical performances of the single MS–SOFC with SBSC–SSZ cathode
measured at 550–700 °C are shown in Fig. 3.15. The MPD is 1.25, 0.92, 0.61 and
Fig. 3.14 a Thermal cycle stability of the cathode polarization resistance, b Impedance spectra of
the cathode polarization resistance measured before and after the 35 cycles. Reproduced with
permission from Ref. [26]. Copyright 2014, Elsevier
3.3 Results and Discussion
73
Fig. 3.15 Electrochemical characteristics of the single cell measured at 550–700 °C: a I–P–V
characteristics, b Impedance spectra. Reproduced with permission from Ref. [26]. Copyright 2014,
Elsevier
0.39 W cm−2 at 700, 650, 600 and 550 ºC, respectively (Fig. 3.15a). In comparison
to those of the MS–SOFCs with LSFSc–YSZ cathodes reported in Chap. 2
(0.40 W cm−2 at 600 ºC), the power densities obtained here show a significant
increase, especially at the low temperatures. The performance improvement should
be caused by the following reasons: (1) the higher oxide ionic conductivity of SSZ
electrolyte than that of YSZ; (2) the high catalytic activity of the SBSC cathode;
(3) the low calcining temperature which helps to maintain a small particle size of
the infiltrated cathode. Figure 3.15b shows the Nyquist plots of the impedance data
obtained at open circuits for the single cell. The pure ohmic resistance is 0.14, 0.22,
0.34 and 0.52 X cm2 and the combined interfacial polarization resistance is 0.11,
0.13, 0.17 and 0.32 X cm2 at 700, 650, 600 and 550 ºC, respectively.
Long-term stability of the MS–SOFC was measured at 550 ºC and 0.46 A cm−2.
As shown in Fig. 3.16, during the 310 h durability test, no degradation is found. In
comparison, the infiltrated SBSC–SSZ cathode was not stable when measured at
Fig. 3.16 Stability of the
single cell measured
at 550 °C
74
3 Fabrication and Investigation of Low-Temperature MS–SOFCs
Fig. 3.17 OCV and MPD of the MS–SOFC retention upon 8 thermal cycles
620 ºC (Fig. 3.5). Thus, reducing the operating temperature would help to increase
the stability of the MS–SOFC with infiltrated electrodes.
The thermal shock resistance of the MS–SOFCs was investigated between 100
and 600 °C at a rate of 10 °C min−1. Figure 3.17 shows the OCV and MPD retention
of the MS–SOFCs upon 8 thermal cycles. Despite the slight decrease of the OCV, no
obvious degradation in the MPD is found. The stable electrochemical performances
of the MS–SOFC are also reflected in the I–P–V curves and the Nyquist plots of the
impedance data of the MS–SOFC before and after the thermal cycling (Fig. 3.18).
The good thermal shock resistance of the MS–SOFC is consistent with the high
thermal cycling stability of the SBSC–SSZ cathode shown in Fig. 3.14.
Fig. 3.18 Electrochemical characteristics of the MS–SOFC before and after the thermal cycles:
a I–P–V characteristics, b Impedance spectra
3.4 Conclusions
3.4
75
Conclusions
In this chapter, to improve the electrochemical performances of MS–SOFCs
operated at low temperatures. Infiltrated Co based cathodes were fabricated and the
polarizations resistances together with the degradation mechanisms were investigated. Low-temperature MS–SOFCs with SBSC–SSZ cathodes were developed
and the electrochemical performances were evaluated. The specific conclusions are
listed as follows:
1. Infiltrated LSC–SSZ, LSCF–SSZ and SBSC–SSZ cathodes were prepared and
the polarization resistances were 0.084, 0.140 and 0.054 X cm2, respectively
(700 °C). Long-term stability test measured at 620 °C shown continuous
increases of both the ohmic and the polarization resistances. The degradation
mechanism was the morphological change of the infiltrated particles, not the
solid state reaction.
2. Low-temperature MS–SOFCs with SBSC–SSZ cathodes were developed and
the MPD was 1.25, 0.92, 0.61 and 0.39 W cm−2 at 700, 650, 600 and 550 ºC,
respectively when measured at 700 °C and no degradation was found during the
310 h measurement (550 °C). Both of the infiltrated SBSC–SSZ cathode and
single MS–SOFC exhibited a high thermal shock resistance.
References
1. Suzuki T, Hasan Z, Funahashi Y, Yamaguchi T, Fujishiro Y, Awano M (2009) Impact of
anode microstructure on solid oxide fuel cells. Science 325:852–855
2. Lee JG, Park JH, Shul YG (2014) Tailoring gadolinium-doped ceria-based solid oxide fuel
cells to achieve 2 W cm−2 at 550 °C. Nat Commun 5:4045
3. Han F, Mücke R, Van Gestel T, Leonide A, Menzler NH, Buchkremer HP et al (2012) Novel
high-performance solid oxide fuel cells with bulk ionic conductance dominated thin-film
electrolytes. J Power Sources 218:157–162
4. Shao ZP, Haile SM (2004) A high-performance cathode for the next generation of solid-oxide
fuel cells. Nature 431:170–173
5. Wachsman ED, Lee KT (2011) Lowering the temperature of solid oxide fuel cells. Science
334:935–939
6. Zhan Z, Bierschenk DM, Cronin JS, Barnett SA (2011) A reduced temperature solid oxide
fuel cell with nanostructured anodes. Energy Environ Sci 4:3951–3954
7. Zhan Z, Han D, Wu T, Ye X, Wang S, Wen T et al (2012) A solid oxide cell yielding high
power density below 600 °C. RSC Adv 2:4075–4078
8. Ullmann H, Trofimenko N, Tietz F, Stöver D, Ahmad-Khanlou A (2000) Correlation between
thermal expansion and oxide ion transport in mixed conducting perovskite-type oxides for
SOFC cathodes. Solid State Ionics 138:79–90
9. Kim JD, Kim GD, Moon JW, Park YI, Lee WH, Kobayashi K et al (2001) Characterization
of LSM-YSZ composite electrode by ac impedance spectroscopy. Solid State Ionics
143:379–389
10. Sholklapper TZ, Kurokawa H, Jacobson C, Visco S, De Jonghe L (2007) Nanostructured
solid oxide fuel cell electrodes. Nano Lett 7:2136–2141
76
3 Fabrication and Investigation of Low-Temperature MS–SOFCs
11. Fukunaga H, Koyama M, Takahashi N, Wen C, Yamada K (2000) Reaction model of dense
Sm0.5Sr0.5CoO3 as SOFC cathode. Solid State Ionics 132:279–285
12. Tucker MC, Lau GY, Jacobson CP, DeJonghe LC, Visco SJ (2008) Stability and robustness
of metal-supported SOFCs. J Power Sources 175:447–451
13. Han D, Liu Y, Wang S, Zhan Z (2014) Enhanced performance of solid oxide fuel cell
fabricated by a replica technique combined with infiltrating process. Int J Hydrogen Energy
39:13217–13223
14. Liu Y, Chen K, Zhao L, Chi B, Pu J, Jiang SP et al (2014) Performance stability and
degradation mechanism of La0.6Sr0.4Co0.2Fe0.8O3−d cathodes under solid oxide fuel cells
operation conditions. Int J Hydrogen Energy 39:15868–15876
15. Xu Q, Huang D-p, Zhang F, Chen W, Chen M, Liu H-x (2008) Structure, electrical
conducting and thermal expansion properties of La0.6Sr0.4Co0.8Fe0.2O3−d–Ce0.8Sm0.2O2−d
composite cathodes. J Alloys Compd. 454:460–465
16. Nie L, Liu M, Zhang Y, Liu M (2010) La0.6Sr0.4Co0.2Fe0.8O3−d cathodes infiltrated with
samarium-doped cerium oxide for solid oxide fuel cells. J Power Sources 195:4704–4708
17. Liu Y, Wang F, Chi B, Pu J, Jian L, Jiang SP (2013) A stability study of impregnated LSCF–
GDC composite cathodes of solid oxide fuel cells. J Alloy Compd 578:37–43
18. Zhao E, Jia Z, Liu X, Gao K, Huo H, Xiong Y (2014) Stability of nanorod-structured
La0.8Sr0.2Co0.2Fe0.8O3−d–Gd0.2Ce0.8O1.9 composite cathodes for intermediate temperature
solid oxide fuel cells. Ceram Int 40:14891–14898
19. Shah M, Voorhees PW, Barnett SA (2011) Time-dependent performance changes in
LSCF-infiltrated SOFC cathodes: the role of nano-particle coarsening. Solid State Ionics
187:64–67
20. Han D, Wu H, Li J, Wang S, Zhan Z (2014) Nanostructuring of SmBa0.5Sr0.5Co2O5+d
cathodes for reduced-temperature solid oxide fuel cells. J Power Sources 246:409–416
21. Meng X, Han D, Wu H, Li J, Zhan Z (2014) Characterization of SrFe0.75Mo0.25O3−d–
La0.9Sr0.1Ga0.8Mg0.2O3−d composite cathodes prepared by infiltration. J Power Sources
246:906–911
22. Kim JH, Cassidy M, Irvine JT, Bae J (2009) Advanced electrochemical properties of LnBa0.
5Sr0. 5Co2O5+d (Ln = Pr, Sm, and Gd) as cathode materials for IT-SOFC. J Electrochem Soc
156:B682–B689
23. Ding X, Kong X, Wu H, Zhu Y, Tang J, Thong Y (2012) SmBa0.5Sr0.5Cu2O5+d and
SmBa0.5Sr0.5CuFeO5+d layered perovskite oxides as cathodes for IT-SOFCs. Int J Hydrogen
Energy 37:2546–2551
24. Li J, Wang S, Liu R, Wang Z, Qian JQ (2008) Electrochemical performance of
(Bi2O3)(1–x)(Er2O3)(x)-Ag composite material for intermediate temperature solid oxide fuel
cell cathode. Solid State Ionics 179:1597–1601
25. Vohs JM, Gorte RJ (2009) High-Performance SOFC Cathodes Prepared by Infiltration. Adv
Mater 21:943–956
26. Zhou Y, Han D, Yuan C, Liu M, Chen T, Wang S et al (2014) Infiltrated
SmBa0.5Sr0.5Co2O5+d cathodes for metal-supported solid oxide fuel cells. Electrochim Acta
149:231–236
27. Zhan W, Zhou Y, Chen T, Miao G, Ye X, Li J et al (2015) Long-term stability of infiltrated
La0.8Sr0.2CoO3−d, La0.58Sr0.4Co0.2Fe0.8O3−d and SmBa0.5Sr0.5Co2.0O5+d cathodes for low
temperature solid oxide fuel cells. Int J Hydrogen Energy 40:16532–16539
Chapter 4
Fabrication and Investigation
of MS–SOFCs with a Symmetric
Configuration
4.1
Introduction
In the previous chapters, by optimizing materials and structures of the MS–SOFCs,
the operating temperatures of the single cells have been reduced from intermediatetemperatures (600–800 °C) to low temperatures (<600 °C), at which the oxidation
rate of the porous stainless steels can be greatly reduced. In this chapter, to simplify
the structure and enhance the performance of the MS–SOFCs, a symmetric configuration with “anode materials infiltrated 430L-electrolyte-cathode materials
infiltrated 430L” was developed. This cell configuration with stainless steel in both
the anode and cathode sides greatly reduces the dependency of expensive rare earth
oxides in the SOFC technology. In this symmetric configuration, the 430L has three
main functions: (1) acting as the supporting layer to provide mechanical strength for
the whole cell; (2) acting as the backbone of the electrodes to support the infiltrated
materials; (3) acting as the electronic conductor and current collector. Due to the
low cost, high mechanical strength and excellent thermal and electrical conductivity
of the stainless steel, development of the symmetric configuration would significantly reduce the cost of the raw materials and the fabrication process, enhance the
mechanical strength, and improve the thermal shock resistance and weldability of
the SOFCs.
In this chapter, infiltrated SmBa0.5Sr0.5Co2.0O5+d (SBSC)-430L electrode was
proposed and evaluated as the possible cathode for MS–SOFCs. A symmetric
MS–SOFC with the configuration of “infiltrated Ni-Ce0.8Sm0.2O2−d (SDC)-430L
anode | scandia stabilized zirconia (SSZ) electrolyte | infiltrated SBSC-430L
cathode” was developed. Reaction kinetics of the cathode and the electrochemical
performance of the MS–SOFCs were also investigated.
© Springer Nature Singapore Pte Ltd. 2018
Y. Zhou, Study on Fabrication and Performance of Metal-Supported Solid Oxide
Fuel Cells, Springer Theses, https://doi.org/10.1007/978-981-10-6617-7_4
77
4 Fabrication and Investigation of MS–SOFCs …
78
4.2
4.2.1
Experimental Section
Fabrication of Symmetric and Single Cells
The new SOFC architecture was based upon tri-layer structures of “porous 430L |
dense SSZ | porous 430L”, which were fabricated using the tape casting, tape
lamination and co-sintering techniques. Commercial 430L stainless steel powder
(−400 mesh, Jing-Yuan Powder Material Corporation, China), (ZrO2)0.89(Sc2O3)0.1
(CeO2)0.01 (Tosoh Corporation, Japan) and ammonium oxalate hydrate, acrylic
resin and methyl benzoate (Sinopharm Chemical Reagent Corporation, China) were
used as the starting materials. Ammonia oxalate hydrate was added as the fugitive
material for the two porous layers. The tri-layer structures were produced by
co-sintering the laminated green tapes at 1325 °C for 4 h in 5% H2–95% N2.
To fabricate single MS–SOFCs with tri-layer structures of “porous 430L | dense
SSZ | porous 430L”, nano-scale catalysts were coated inside the pore walls of the
porous 430L scaffolds: Ni–SDC at the anodes and SBSC at the cathodes. The
mixed nitrate solutions of Ni(NO3)2, Sm(NO3)3 and Ce(NO3)3 at stoichiometric
ratios (SDC:Ni = 80:20 wt) were infiltrated into one of the porous 430L scaffolds
while the opposite scaffolds were infiltrated with an aqueous solution containing
stoichiometric amounts of Sm(NO3)3, Ba(NO3)2, Sr(NO3)2 and Co(NO3)2, followed
by calcinations at 650 °C in air for 2 h to prevent oxidation of 430L alloys. Note
that citric acid (CA) was added as the chelating agent into the SBSC precursor
solution with the molar ratio of CA to the total cations at 1.5. The volume of
infiltration solutions was controlled by a micro-liter syringe, and a single
infiltration/firing cycle produced a catalyst loading of 2–3 wt%. The ultimate
loadings of the infiltrated catalysts in the porous 430L scaffolds were tailored by
varying the number of infiltration/firing cycles. Different from the single MS–
SOFCs, symmetric anode and cathode cells were obtained by infiltrating both
porous 430L scaffolds with the same catalysts, i.e., Ni–SDC for anodes and SBSC
for cathodes.
4.2.2
Material Characterizations
The cell structure after measurements was examined using the scanning electron
microscopy (SEM) in a Hitachi S-4800-II and FEI Magellan 400 microscopes.
Phase compositions of SBSC and Ni–SDC powders, as synthesized by distillating
and calcining the infiltration solutions, were identified at room temperature using a
Rigaku XRD diffractometer with monochromatic CuKa radiation. The morphologies of reduced Ni–SDC and as-synthesized SBSC catalysts were examined using
the transmission electron microscopy on a JEM-2100F microscope.
4.2 Experimental Section
4.2.3
79
Electrochemical Measurements
For electrochemical measurements, silver ink was applied on the electrode surface
as the current collector, and silver wires were attached as the current and voltage
leads. Single MS–SOFCs were tested at 500–600 °C with 97% H2–3% H2O in the
anodes and air in the cathodes both at 100 sccm. Current–voltage curves and
electrochemical impedance spectra were obtained using an IM6 Electrochemical
Workstation (ZAHNER, Germany). The frequency range for impedance measurement was 0.1 Hz to 1 MHz. For symmetric cells, impedance data were collected in
a uniform atmosphere—97% H2–3% H2O for Ni–SDC–430L anodes and dry air for
SBSC–430L cathodes.
4.3
4.3.1
Results and Discussion
Investigation of Infiltrated SBSC–430L Cathodes
To verify the oxidation resistance of the porous 430L layer fabricated by the tape
casting and sintering method, the 430L layer was thermally treated in air at 600 °C
and the weight variation was recorded. As shown in Fig. 4.1, high oxidation
resistance of the porous 430L scaffolds is confirmed by negligible weight gains [1].
Figure 4.2 shows the polarization resistances of the infiltrated SBSC–430L
cathodes measured at 650 °C in air with the SBSC loadings ranging from 6 to
20 wt%. It can be found that the composite cathode with the SBSC loading of
16 wt% exhibits the lowest polarization resistance. Both the polarization resistances
of SBSC loadings lower and higher than 16 wt% are much larger than that of
the 16 wt% loading, which can be explained by the three-phase boundary (TPB)
dependence of the infiltrated loadings [2].
Nyquist plots of the impedance data of the polarization resistances of infiltrated
SBSC–430L cathodes measured at 500–650 °C in air with the SBSC loading of
16 wt% are shown in Fig. 4.3. The polarization resistances of the SBSC–430L
cathodes are 0.040, 0.093, 0.141 and 0.185 X cm2 when measured at 650, 600,
Fig. 4.1 Weight variation of
the porous 430L scaffolds as
thermally treated in air at
600 °C. Reproduced with
permission from Ref. [1].
Copyright 2015, Wiley-VCH
80
4 Fabrication and Investigation of MS–SOFCs …
Fig. 4.2 a Nyquist plots of the impedance data of the polarization resistances of SBSC–430L
cathodes with different SBSC loadings, b Effect of SBSC loadings on the polarization resistances
of the SBSC–430L symmetric cathode cells. Reproduced with permission from Ref. [1].
Copyright 2015, Wiley-VCH
Fig. 4.3 Impedance spectra
of the polarization resistances
of the SBSC–430L symmetric
cathode cell measured at
500–650 °C
575 and 550 °C, respectively. The promising results obtained here are comparable
with those of the all-ceramic cathodes, e.g., La0.6Sr0.4Co0.2Fe0.8O3−d, Ba0.5Sr0.5
Co0.8Fe0.2O3−d and Sm0.5Sr0.5CoO3−d [3–5].
Cr poisoning is one of the potential issues using stainless steels in the cathode
side, which will lead to the electrochemical deactivation of the cathode [6–8].
Figure 4.4 shows the stability of the polarization resistance of the SBSC–430L
cathode and no degradation of the resistance is observed during the 300 h
heat-treatment at 500 °C. It illustrates that the volatilization of Cr and the reaction
between Cr and SBSC can be ignored at such a low operating temperature.
Fig. 4.4 Stability of the
polarization resistance of the
SBSC–430L symmetric
cathode cell. Reproduced with
permission from Ref. [1].
Copyright 2015, Wiley-VCH
4.3 Results and Discussion
81
Fig. 4.5 Schematic illustration of the physical-chemical processes occurring in the MS–SOFC.
Reproduced with permission from Ref. [1]. Copyright 2015, Wiley-VCH
Schematic illustration of the physical-chemical processes occurring in the
MS–SOFC is shown in Fig. 4.5. The oxygen reduction reaction (ORR) occurs in
the SBSC–430L cathode, then oxide ions are transported by the interconnected
SBSC coatings to the adjacent SSZ electrolyte. On the anodes, oxide ions diffuse
through the bulk SDC coatings to the triple-phase boundaries (TPBs), where SDC,
Ni, and H2 come into contact to facilitate the hydrogen oxidation reaction.
Based upon the Alder–Lane–Steele model [9], the convoluted chemical contributions to the cathode impedance of oxygen surface exchange and solid-state diffusion are reflected by a Gerischer-type response in the EIS data, which can be
expressed as:
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1
ð4:1Þ
ZH ¼ Rchem
1 þ jwtchem
where Rchem and tchem are the characteristic resistance and time constant, respectively. Prior reports have shown that the charge transfer impedance appeared at
higher frequencies and was independent of oxygen partial pressures (Po2) while the
Po2-sensitive Gerischer impedance was commonly observed at lower frequencies
[9]. In order to gain insights into the oxygen reduction mechanism, impedance
measurements were performed on symmetric cathode cells under varied oxygen
partial pressures and temperatures. The equivalent circuits used for EIS data fitting
are shown in Fig. 4.6, where L represents the conductance due to the connecting.
Fig. 4.6 Schematics of the equivalent circuit used for impedance spectra fitting. Reproduced with
permission from Ref. [1]. Copyright 2015, Wiley-VCH
82
4 Fabrication and Investigation of MS–SOFCs …
Fig. 4.7 a Representative
impedance spectra of
symmetric cathode cells
measured under varied
oxygen partial pressures,
b Oxygen partial pressure
dependence of the charge
transfer resistances (Rct), the
Gerischer resistances (Rchem),
and the total polarization
resistances (R p,c).
Reproduced with permission
from Ref. [1]. Copyright
2015, Wiley-VCH
The higher-frequency component was fitted with a charge transfer resistance (Rct)
in parallel with a constant phase element (CPE) while the lower-frequency component was fitted with a Gerischer-type response (H).
Nyquist plots of the impedance data shown in Fig. 4.7a yields the higherfrequency (>1 kHz) charge transfer resistances (Rct) and the lower-frequency
Gerischer resistances (Rchem). Figure 4.7b summarizes the Rct and Rchem values at
varied oxygen partial pressures. Indeed, the Rct values remain largely constant
at 0.18 X cm2 at 525 °C. Note that the cathode polarization resistance and the
oxygen partial pressure follow an exponential correlation as:
RP / Pn
O2
ð4:2Þ
where the n value provides the information on the rate-limiting step in the oxygen
reduction reactions. The obtained n value by linear fitting is n = 0.001 for Rct,
indicating the diffusion of oxide ions from TPBs to the electrolyte together with a
charge transfer process; n = 0.13 for Rchem, suggesting that the Gerischer resistance
is probably correlated with double ionization of the adsorbed oxygen [10]:
2
O
S þ e , OS
ð4:3Þ
Larger Rchem values than Rct at Po2 = 0.2 atm in Fig. 4.7b suggest that ORRs on
the SBSC–430L cathode are more limited by double ionization of adsorbed oxygen
4.3 Results and Discussion
83
Fig. 4.8 Temperature
dependence of the charge
transfer resistances (Rct) and
Gerischer chemical resistance
(Rchem) of symmetrical
cathode cell. Reproduced with
permission from Ref. [1].
Copyright 2015, Wiley-VCH
at 525 °C. Nonetheless, oxide ion charge transfer across the SBSC | SSZ interfaces
becomes dominating at higher temperatures, e.g., 600 °C, due to the larger activation energy of 1.37 eV for Rchem compared with 0.88 eV for Rct (Fig. 4.8).
4.3.2
Investigation of MS–SOFCs with a Symmetric
Configuration
A symmetric MS–SOFC with the configuration of “infiltrated Ni–SDC–430L
anode–SSZ electrolyte-infiltrated SBSC–430L cathode” was developed.
Figure 4.9a shows a typical SEM image of tri-layer structures of “porous 430L |
dense SSZ | porous 430L” before infiltrating. The SSZ electrolyte is fully dense and
typically 15 lm thick. Nanoporous and well interconnected coatings of SDC–Ni
(loading = 10 wt%) and SBSC (loading = 16 wt%) are well coated onto the 430L
backbones (Fig. 4.9b and c).
Figure 4.10 shows the SEM images of the infiltrated electrodes after the electrochemical characterizations of the MS–SOFC. Ni–SDC particles with particle
sizes ranging from 20 to 100 nm and SBSC particles around 50 nm are shown.
Such nano-porous structures with abundant surface areas are benefit for the electrochemical reactions in the electrodes.
Formation of Ni–SDC and SBSC in the coatings was independently confirmed
by X-ray diffraction patterns of powders, synthesized by calcining the infiltrating
solutions at 650 °C for 1 h in air (Fig. 4.11). To mimic the operating condition of
the MS–SOFC, Ni–SDC powder was further heat treated at 650 °C for 1 h in 3%
H2O–97% H2. Pure Ni and SDC phases are obtained while some minor impurities
of BaCoO3 are also observed in SBSC.
Examination of as-synthesized catalysts by transmission electron microscopy
(TEM), with the results illustrated in Fig. 4.12a for Ni–SDC and in Fig. 4.12b for
SBSC, reveals a typical porous aggregate morphology that consist of 10–40 nm
particles. This observation is consistent with the XRD crystalline sizes calculated
by the Debye–Scherrer method—18 nm for SDC, 35 nm for Ni, and 19 nm for
84
4 Fabrication and Investigation of MS–SOFCs …
Fig. 4.9 Cross-sectional SEM images of the: a 430L/SSZ/430L backbone, b Ni–SDC–430L
anode, c SBSC–430L cathode. Reproduced with permission from Ref. [1]. Copyright 2015,
Wiley-VCH
Fig. 4.10 SEM images of the: a Ni–SDC particles, b SBSC particles
SBSC. The selected area electron diffraction (SAED) patterns for SDC (Fig. 4.13a)
and SBSC (Fig. 4.13c) show the presence of diffuse rings along with some Bragg
spots, indicative of a highly defective or polycrystalline nature for these two catalysts. Differently, the SAED patterns for Ni (Fig. 4.13b) suggest that these
nanoparticles are single crystals. The associated characteristic speckle patterns in
Fig. 4.13b should arise from the adjacent smaller single crystals of Ni metals.
Elemental mapping for the TEM images of the Ni–SDC and SBSC particles are
shown in Figs. 4.14 and 4.15, respectively. As shown, the Ni and SDC two phases
can be clearly distinguished with aggregated coarsening Ni particles and dispersed
4.3 Results and Discussion
85
Fig. 4.11 XRD patterns of the: a Ni–SDC powder, b SBSC powder. Reproduced with permission
from Ref. [1]. Copyright 2015, Wiley-VCH
Fig. 4.12 TEM micrographs of the: a Ni–SDC powder, b SBSC powder. Reproduced with
permission from Ref. [1]. Copyright 2015, Wiley-VCH
Fig. 4.13 Selected area electron diffraction (SAED) patterns of the: a SDC particles, b Ni
particles, c SBSC particles. Reproduced with permission from Ref. [1]. Copyright 2015,
Wiley-VCH
86
4 Fabrication and Investigation of MS–SOFCs …
Fig. 4.14 Elemental mapping for the TEM image of the Ni–SDC powder
Fig. 4.15 Elemental mapping for the TEM image of the SBSC powder
small SDC particles (Fig. 4.14). As for SBSC, a homogenous distribution of element Sm and Sr is shown, while the Ba, Co and O tend to form segregations, which
should be caused by the impurities BaCoO3 shown in Fig. 4.11b.
Electrochemical measurements were performed on the single MS–SOFC, i.e.,
Ni–SDC–430L | SSZ | SBSC–430L, with 3% humidified hydrogen fuels in the
anode and dry air oxidants in the cathode at 500–600 °C. Figure 4.16a shows
typical polarization curves of cell voltages and power densities versus current
densities. The open circuit voltage values increase from 1.104 to 1.122 V with
decreasing temperature, and are within 20 mV of the calculated Nernst potentials.
These results suggest that the SSZ electrolytes co-sintered with the porous 430L
scaffolds are gas impermeable, consistent with the SEM observation in Fig. 4.9a.
The maximum power densities (MPDs) measured are 1.02, 0.86, 0.63, 0.41, and
0.27 W cm−2 at 600, 575, 550, 525, and 500 °C, respectively. These values are
very competitive with the common all-ceramic SOFCs, or prior MS–SOFCs with
4.3 Results and Discussion
87
zirconia-based electrolytes, which can be ascribed to good catalytic activities of
nanoscale catalysts and high oxide ionic conductivities of the SSZ electrolyte
[11–16]. For instances, MS–SOFCs with the FeCr alloy supports, Ce0.9Gd0.1O2–d–
Ni infiltrated cermet anodes, Sc2O3 and Y2O3 co-doped ZrO2 electrolytes and
in situ sintered La0.6Sr0.4CoO3–d cathodes produced peak power densities of
1.14 W cm−2 at a higher temperature of 650 °C [15]. Although there are prior
reports of excellent all-ceramic fuel cells that produced power densities of
1 W cm−2 at 500–550 °C, they were obtained with much more conductive
electrolytes such as Gd3+ doped ceria or strontium and magnesium doped lanthanum gallate [4, 17, 18]. Electrochemical impedance spectroscopy (EIS) of the
MS–SOFC measured at OCVs are shown in Fig. 4.16b. The ohmic resistances are
0.12, 0.17 and 0.26 X cm2 and the polarization resistances are 0.19, 0.23 and 0.40 X
cm2 when measured at 600, 550 and 500 °C, respectively. The polarization resistance dominates the cell performance.
To evaluate the respective contribution of the anode and cathode to the whole
polarization resistance of the single MS–SOFC, symmetric anode and cathode cells
were fabricated and the impedance spectra were measured. As shown in Fig. 4.17,
when measured at 600 °C, the polarization resistance of the Ni–SDC–430L anode
and the SBSC–430L cathode is 0.116 and 0.093 X cm2, respectively. Thus, the
Fig. 4.16 a I–P–V characteristics and b Impedance spectra of the MS–SOFC. Reproduced with
permission from Ref. [1]. Copyright 2015, Wiley-VCH
Fig. 4.17 Impedance spectra of the symmetrical cells: a Ni–SDC–430L anode cells, b SBSC–
430L cathode cells. Reproduced with permission from Ref. [1]. Copyright 2015, Wiley-VCH
4 Fabrication and Investigation of MS–SOFCs …
88
Fig. 4.18 The ohmic, anodic
and cathodic polarization
resistances of the MS–SOFC
plotted versus inverse
temperature. Reproduced with
permission from Ref. [1].
Copyright 2015, Wiley-VCH
Fig. 4.19 Stability of the
MS–SOFC measured at
500 °C. Reproduced with
permission from Ref. [1].
Copyright 2015, Wiley-VCH
anode and cathode makes a comparable contribution to the polarization resistance
of the single MS–SOFC.
The ohmic resistance (RO), anodic resistance (RP,A) and cathodic resistance
(RP,C) values at varied temperatures are summarized in Fig. 4.18, showing that the
ohmic losses, the anodic, and cathodic polarizations almost equally contribute to the
internal resistances for the present MS–SOFCs. A higher activation energy for
oxygen reduction than for hydrogen oxidation (1.13 vs. 0.73 eV) suggests that
optimizing the cathode composition and microstructure would be more effective in
further enhancing the fuel cell performance at lower temperatures.
Time dependence of the voltage of the MS–SOFC measured at 500 °C and 0.25
A cm−2 are shown in Fig. 4.19 and no degradation is observed during the 60 h
duration. The stable output power density of the single cell consistent with the
stable polarization resistance of the SBSC–430L cathode shown in Fig. 4.4.
4.4
Conclusions
In this chapter, to simplify the structure, reduce the cost and enhance the performance of the MS–SOFCs, a symmetric configuration was developed. The
structure-performance relationship of the SBSC–430L cathodes, reaction kinetics of
the cathodes and the electrochemical performances of the MS–SOFCs were
investigated. The specific conclusions are listed as follows:
4.4 Conclusions
89
1. Infiltrated SBSC–430L cathodes were prepared and the polarization resistances
were 0.040, 0.093, 0.141 and 0.185 X cm2 when measured at 650, 600, 575 and
550 °C, respectively. A high stability of the cathode has been demonstrated with
no degradation of the polarization resistance observed during the 300 h
heat-treatment at 500 °C. Kinetic studies indicated that the higher frequency
peak of the impedance spectra of SBSC–430L cathode was associated with the
transport of oxide-ions from the TPB to the electrolyte while the lower frequency peak was attributed to the ionization of absorbent oxygen.
2. MS–SOFCs with the symmetric configuration of “Ni–SDC–430L anode, SSZ
electrolyte and SBSC–430L cathode” were fabricated. MPD of the MS–SOFC
were 0.27 and 1.02 W cm−2 when measured at 500 and 600 °C, respectively
and no degradation was found during the 60 h measurement at 500 °C.
References
1. Zhou Y, Wu H, Luo T, Wang J, Shi Y, Xia C et al (2015) A nanostructured architecture for
reduced—temperature solid oxide fuel cells. Adv Energy Mater 5:1500375
2. Zhang Y, Sun Q, Xia C, Ni M (2013) Geometric properties of nanostructured solid oxide fuel
cell electrodes. J Electrochem Soc 160:F278–F289
3. Shao ZP, Haile SM (2004) A high-performance cathode for the next generation of solid-oxide
fuel cells. Nature 431:170–173
4. Zhan Z, Han D, Wu T, Ye X, Wang S, Wen T et al (2012) A solid oxide cell yielding high
power density below 600 °C. RSC Adv 2:4075–4078
5. Zhao F, Wang Z, Liu M, Zhang L, Xia C, Chen F (2008) Novel nano-network cathodes for
solid oxide fuel cells. J Power Sources 185:13–18
6. Montero X, Tietz F, Sebold D, Buchkremer HP, Ringuede A, Cassir M et al (2008)
MnCo1.9Fe0.1O4 spinel protection layer on commercial ferritic steels for interconnect
applications in solid oxide fuel cells. J Power Sources 184:172–179
7. Park E, Taniguchi S, Daio T, Chou J-T, Sasaki K (2014) Influence of cathode polarization on
the chromium deposition near the cathode/electrolyte interface of SOFC. Int J Hydrogen
Energy 39:1463–1475
8. Komatsu T, Chiba R, Arai H, Sato K (2008) Chemical compatibility and electrochemical
property of intermediate-temperature SOFC cathodes under Cr poisoning condition. J Power
Sources 176:132–137
9. Adler S (2000) Limitations of charge-transfer models for mixed-conducting oxygen
electrodes. Solid State Ionics 135:603–612
10. Wang Y, Zhang L, Chen F, Xia C (2012) Effects of doped ceria conductivity on the
performance of La0.6Sr0.4Co0.2Fe0.8O3−d cathode for solid oxide fuel cell. Int J Hydrogen
Energy 37:8582–8591
11. Han F, Mücke R, Van Gestel T, Leonide A, Menzler NH, Buchkremer HP et al (2012) Novel
high-performance solid oxide fuel cells with bulk ionic conductance dominated thin-film
electrolytes. J Power Sources 218:157–162
12. Zhan Z, Bierschenk DM, Cronin JS, Barnett SA (2011) A reduced temperature solid oxide
fuel cell with nanostructured anodes. Energy Environ Sci 4:3951–3954
13. Tucker MC (2010) Progress in metal-supported solid oxide fuel cells: a review. J Power
Sources 195:4570–4582
14. Zhou Y, Ye X, Li J, Zhan Z, Wang S (2014) Metal-supported solid oxide fuel cells with a
simple structure. J Electrochem Soc 161:F332–F336
90
4 Fabrication and Investigation of MS–SOFCs …
15. Klemenso T, Nielsen J, Blennow P, Persson AH, Stegk T, Christensen BH et al (2011) High
performance metal-supported solid oxide fuel cells with Gd-doped ceria barrier layers.
J Power Sources 196:9459–9466
16. Kim KH, Park YM, Kim H (2010) Fabrication and evaluation of the thin NiFe supported solid
oxide fuel cell by co-firing method. Energy 35:5385–5390
17. Han D, Wu H, Li J, Wang S, Zhan Z (2014) Nanostructuring of SmBa0.5Sr0.5Co2O5+d
cathodes for reduced-temperature solid oxide fuel cells. J Power Sources 246:409–416
18. Lee JG, Park JH, Shul YG (2014) Tailoring gadolinium-doped ceria-based solid oxide fuel
cells to achieve 2 W cm−2 at 550 °C. Nat Commun 5:4045
Chapter 5
Summary and Outlook
5.1
Summary
This thesis aims to develop novel metal-supported solid oxide fuel cells (MS–
SOFCs) with low-cost and mechanical robust stainless steels (SS430L) replacing
the ceramic materials as the supports for SOFCs. In order to solve the issues during
the cell fabrication and operation processes, and enhance the electrochemical performance and stability of MS–SOFCs, a “tape casting-sintering-infiltrating” method
and a “micro-nano” structure were developed. Besides, the structure-performance
relationship of the electrodes, reaction kinetics of the electrodes and degradation
mechanisms of the fuel cells were also investigated. The main conclusions of this
thesis are summarized as follows.
Fabrication and Investigation of Intermediate-Temperature MS–SOFCs
To enhance the performances of intermediate-temperature MS–SOFCs (600–800 °C),
La0.6Sr0.4Fe0.9Sc0.1O3−d (LSFSc) cathode and Ni-based anode materials were applied
by the infiltration method to reduce the polarization resistance of cathode and anode,
respectively. When measured at 650 °C, polarization resistances of the infiltrated
LSFSc–YSZ cathode, Ni–430L anode, SDC–430L anode and Ni–SDC–430L anode
were 0.160, 2.2, 0.233 and 0.112 X cm2, respectively. Kinetic studies indicated that
the low frequency peak of the impedance spectra of the SDC–430L anode was
associated with the extraction of surface lattice oxygen of SDC while the high frequency peak was attributed to the transport of oxide-ions within the SDC coating.
Maximum power densities (MPDs) of the MS–SOFC with Ni–SDC–430L anode
were 1.23, 0.92 and 0.40 W cm−2 when measured at 800, 700 and 600 °C, respectively. A degradation rate of 1.3% kh−1 was shown when measured at 650 ºC for
1500 h. The degradation rate was faster at higher operation temperatures and larger
current densities and the degradation was mainly caused by the morphological change
of the anode.
© Springer Nature Singapore Pte Ltd. 2018
Y. Zhou, Study on Fabrication and Performance of Metal-Supported Solid Oxide
Fuel Cells, Springer Theses, https://doi.org/10.1007/978-981-10-6617-7_5
91
92
5 Summary and Outlook
Fabrication and Investigation of Low-Temperature MS–SOFCs
To enhance the performances of low-temperature MS–SOFCs (<600 °C), infiltrated
La0.8Sr0.2CoO3−d (LSC)-scandia stabilized zirconia (SSZ), La0.58Sr0.4Co0.2Fe0.8O3−d
(LSCF)–SSZ and SmBa0.5Sr0.5Co2.0O5+d (SBSC)–SSZ cathodes were investigated.
Polarization resistances of the infiltrated LSC–SSZ, LSCF–SSZ and SBSC–SSZ
cathodes were 0.084, 0.140 and 0.054 X cm2, respectively when measured at 700 °C.
Long-term stability test measured at 620 °C shown continuous increases of both the
ohmic and the polarization resistances of the cathodes. The degradation mechanism
was the morphological change of the infiltrated particles, not the solid state reaction.
Low-temperature MS–SOFCs with SBSC–SSZ cathodes were developed and the
MPD were 1.25, 0.92, 0.61 and 0.39 W cm−2 at 700, 650, 600 and 550 ºC, respectively
when measured at 700 °C and no degradation was found during the 310 h measurement (550 °C). Both of the infiltrated SBSC–SSZ cathode and single MS–SOFC
exhibited a high thermal shock resistance.
Fabrication and Investigation of MS–SOFCs with a Symmetric Configuration
To simplify the structure, reduce the cost and enhance the performance of the MS–
SOFCs, a symmetric configuration was developed. Infiltrated SBSC–430L cathodes
were prepared and the polarization resistances were 0.040, 0.093, 0.141 and 0.185 X
cm2 when measured at 650, 600, 575 and 550 °C, respectively. A high stability of the
cathode has been demonstrated with no degradation of the polarization resistance
observed during the 300 h heat-treatment at 500 °C. Kinetic studies indicated that the
higher frequency peak of the impedance spectra of SBSC–430L cathode was
associated with the transport of oxide-ions from the three-phase boundary (TPB) to
the electrolyte while the lower frequency peak was attributed to the ionization of
absorbent oxygen. MS–SOFCs with the symmetric configuration of “Ni–SDC–430L
anode, SSZ electrolyte and SBSC–430L cathode” were fabricated. MPD of the MS–
SOFC were 0.27 and 1.02 W cm−2 when measured at 500 and 600 °C, respectively
and no degradation was found during the 60 h measurement at 500 °C.
5.2
Outlook
In this thesis, MS–SOFCs have been successfully fabricated and promising electrochemical performances and long-term stabilities have been demonstrated. The
structure-performance relationship of the electrodes, reaction kinetics of the electrodes and degradation mechanisms of the MS–SOFCs were also investigated.
Despite the achievements, future work, e.g., the application of in-situ characterization techniques, and optimization of the structures and materials are still needed
to further investigate the structure and reaction kinetics of the electrode materials,
improve the electrochemical performances and stabilities of the fuel cells, and
expand the application areas of the MS–SOFCs.
5.2 Outlook
93
(1) In-situ characterization techniques, e.g., in-situ Raman spectroscopy, X-ray
photoelectron spectroscopy (XPS) and absorption near edge structure (XANES)
are needed to gain a deeper insight into the structure evolution of the infiltrated
nano materials, thus to illustrate the electrochemical activity and the degradation mechanisms of the electrodes.
(2) Further optimizing the materials and structures of the MS–SOFCs, e.g., the
application of highly conductive electrolytes to improve the electrochemical
performances and stabilities of the MS–SOFCs.
(3) Expanding the application areas of the MS–SOFCs, e.g., using hydrocarbons as
the fuels and applying the MS–SOFCs for water electrolysis.
Документ
Категория
Без категории
Просмотров
0
Размер файла
5 111 Кб
Теги
978, 981, 6617
1/--страниц
Пожаловаться на содержимое документа