- Email: [email protected]
Influence of pore former on electrochemical performance of fuel-electrode supported SOFCs manufactured by aqueous-based tape-casting
Energy 115 (2016) 149e154
Contents lists available at ScienceDirect
Energy journal homepage: www.elsevier.com/locate/energy
Influence of pore former on electrochemical performance of fuel-electrode supported SOFCs manufactured by aqueous-based tape-casting Juan Zhou b, *, Qinglin Liu b, Lan Zhang b, Zehua Pan a, b, Siew Hwa Chan a, b, ** a b
School of Mechanical and Aerospace Engineering, Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798, Singapore Energy Research Institute at NTU (ERI@N), Nanyang Technological University, 1 CleanTech Loop #06-04, Singapore 637141, Singapore
a r t i c l e i n f o
a b s t r a c t
Article history: Received 15 December 2015 Received in revised form 22 July 2016 Accepted 27 August 2016
The microstructure of a fuel electrode has a significant influence on the whole performance of fuelelectrode supported solid oxide fuel cells (SOFCs) fabricated by aqueous-based tape casting. While in the aqueous-based tape casting process, the fuel electrode porosity, which plays a key role in the final fuel electrode microstructure, mainly comes from the pore former (potato starch in this case). Different contents of starch are added into fuel electrode slurries. When the starch content is 2.5%wt, the cells show the best performance. After one thermal cycle and discharging at a constant voltage of 0.7 V and temperature of 800 C, the peak power density has reached ~1263 mW cm2 with humidified H2 as the fuel and air as the oxidant. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Solid oxide fuel cells (SOFCs) Aqueous-based tape casting Pore former Starch
1. Introduction Solid oxide fuel cells (SOFCs) technology is a cutting-edge technology for directly converting the chemical energy of the fuel into the electrical energy and heat by means of the electrochemical reactions [1,2]. SOFCs technology has many advantages compared with the conventional power plants, such as high efficiency, fuel flexibility, environmentally friendly, low noise level, and etc. [3e5]. SOFCs are mainly divided into four categories, namely the fuel electrode-, air electrode-, electrolyte-, and structural-supported SOFCs [6,7]. Different types of supports have different characteristics, hence exhibiting both advantages and disadvantages. However, the fuel electrode-supported SOFC is considered to be the most mature one compared to the other three [8]. The supporting
List of abbreviation and nomenclature: SOFCs, solid oxide fuel cells; TPB, three phase boundary; OCV, open circuit voltages; I-V, current-voltage; I-P, current-power density; SEM, scanning electron microscope; RU, ohmic resistances; RT, total resistances. * Corresponding author. ** Corresponding author. School of Mechanical and Aerospace Engineering, Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798, Singapore. E-mail addresses: [email protected] (J. Zhou), [email protected] (S.H. Chan). http://dx.doi.org/10.1016/j.energy.2016.08.093 0360-5442/© 2016 Elsevier Ltd. All rights reserved.
fuel electrode provides the mechanical strength to the whole cell, and offers the cell with high electronic conductivity and porosity. The microstructure of the supporting fuel electrode is crucial to the performance of the entire cell [9e12] as gas permeability and reactions strongly depend on microstructural parameters such as porosity, particle size and three phase boundary length [13e15]. Porous nickel-yttria stabilized zirconia (Ni-YSZ) is currently the most common fuel electrode material for SOFCs [16e18]. During the fabrication process, the starting materials of fuel electrode (NiO and YSZ mixture) are prepared by high temperature processing. Then NiO-YSZ is reduced to metallic Ni-YSZ when the fuel electrode is exposed to the fuel environment during the cell operation. In the process of reduction from NiO to Ni, the volume of particles shrunk, which creates some amount of porosity, but this amount of porosity is insufficient to meet the requirements of gas permeability, electrochemical reactions and steam removal. Hence, the pore former is added to increase the porosity. However, if the porosity is lower and higher than the optimal value, the area of the three phase boundary region would not be maximised, which infers that it is crucial to determine the best amount value of porosity with the optimized amount of pore former added into the fuel electrode slurry. Tape casting is a well-known shaping technology for producing large area, thin, uniform and flat ceramic tapes. This technology (organic-based tape casting) is more popular in SOFCs
150
J. Zhou et al. / Energy 115 (2016) 149e154
Table 1 The formula of different fuel electrode slurries with different amounts of poreformer. Chemicals
The ratio of the fuel electrode layer (%)
NiO YSZ Potato starch Poly acrylic acid (PAA) De-ionised water Poly vinyl alcohol, 87e89% partially hydrolysed Poly ethylene glycol (PEG)(Mw 380e420) Glycerol 1-Octanol 2,4,7,9-Tetramethyl-5-decyne-4,7-diol ethoxylate
50 50 1.5, 2, 2.5, 3, 3.5 2 75 10.5 14 14 1.2 0.75
manufacturing. However, organic-based tape casting often involves the use of toxic solvents and hazardous additives, which not only increases the manufacturing cost, but is also harmful to human health and the environment. For this reason, there are increasingly more attentions paid to development of aqueous-based tape casting [19e21]. While in a stable aqueous-based tape casting slurry system, the porosity of fuel electrode is mainly determined by the amount of pore former. There are several papers studied the effect of different pore formers on the properties of the fuel electrode [22e28], while only a handful of literature have been reported with regard to the influence of pore formers on the morphology in the aqueous tape casting [29]. The aim of this work is to study the influence of the pore former (potato starch in this paper) on the microstructure of fuel electrode and its associated electrochemical performance in the aqueous-based tape casting. 2. Experimental 2.1. Fabrication Firstly, the aqueous co-tape casting and co-sintering were used to fabricate fuel-electrode supported half cells. Commercial NiO (J. T. Baker, US) and Zr0.92Y0.08O2-d (YSZ, TOSOH, Japan) were used for preparing the fuel-electrode layer, and the YSZ powder was used to prepare the electrolyte layer. In the aqueous-based tape casting process, potato starch powder (pore former), polyacrylic acid (disperse agent), ammonia (electrostatic dispersant), hydrolysed polyvinyl alcohol (the binder), glycerol and polyethylene glycol (plasticizer), 1-octanol (de-foamer) and 2,4,7,9-tetramethyl-5-decyne-4,7-diol ethoxylate (surfactant) were added. Details of the fabrication process can be found in our
Fig. 2. The picture of (a) 10 cm 10 cm half cell and (b) the button cell.
previous paper [30]. After the NiO þ YSZ/YSZ half-cell was gotten, the La0.6Sr0.4Co0.2Fe0.8O3þd(LSCF)/Ce0.9Gd0.1O2þd (CGO) composite air-electrode with a weight ratio of 1:1 was screen printed on it. To get a stable slurry system, the disperse agent, the electrostatic dispersant, the binder, the plasticizer, the de-foamer and the surfactant need to be in a fixed formula, even though each they can affect the porosity separately, while pore former can be more flexible. Hence, it is much easier to study the influence of the electrode microstructure by changing the amount of pore-former. Table 1 shows different amounts of pore-former used in the fuel electrode slurries with the weight ratio of 1.5, 2, 2.5, 3, 3.5%, respectively. Finally, the 10 cm by 10 cm single cells were fabricated. Fig. 1 shows the cross-section micrographs of the cell with
Fig. 1. The microstructure of single cell with the weight ratio of 2.5%: (a) the low magnification, and (b) high magnification SEM micrograph of the cell's cross-sectional view.
J. Zhou et al. / Energy 115 (2016) 149e154
151
700 C with humidified H2 as the fuel and air as the oxidant. The flow rate of the humidified hydrogen was set at 50 ml min1. The polarisation curve and electrochemical impedance spectrum were obtained using an Autolab PG30/FRA system (Eco Chimie, Netherlands) in the frequency range of 0.1 Hze100 KHz with an excitation potential of 10 mV. A four-probe configuration was adopted in the electrochemical testing. The microstructure of the ceramic samples was characterized by a scanning electron microscope (SEM, JEOL 5600, Japan).
3. Results and discussion
Fig. 3. Typical polarisation curves of voltage and power density versus the current density for different single cells with different amount of pore former at 800 C.
Fig. 4. The impedance spectra of different cells at 800 C.
the weight ratio of 2.5%. 2.2. Measurements To study the influence of different amounts of pore former on electrochemical performance much more effectively, button cells with a diameter of 2.5 cm were cut off from the large area samples. Fig. 2(a) shows the picture of 10 cm 10 cm cell and (b) button cells. The effective area of the air electrode was ~0.5 cm2, and the ceramic paste was used as sealer. The cell performance and electrochemical impedance were measured from 800 C down to
Based on different amounts of pore former in the green fuelelectrode, the cells with 1.5, 2, 2.5, 3, 3.5 wt% potato starches are named as Cell-1.5, Cell-2, Cell-2.5, Cell-3 and Cell-3.5, respectively. Compositions of the cells are all NiO þ YSZ/YSZ/LSCF þ CGO. Fig. 3 shows typical polarisation curves of voltage and power density versus the current density for different single cells with different amounts of pore former at 800 C. The Cell-2.5 and Cell-3 have achieved better electrochemical performance with the peak power density of 560.8 and 570.2 mW cm2 respectively than others. The open circuit voltages (OCV) values are 1.08, 1.1, 1.02, 1.09 and 1.09 V for Cell-1.5, Cell-2, Cell-2.5, Cell-3 and Cell-3.5, respectively. The OCV of Cell-2.5 is slightly lower, while other cells are in good agreement with the theoretical values calculated from the Nernst equation. Fig. 4 shows the impedance spectra of the five cells operated at 800 C. The ohmic resistances (RU) of the five cells are 0.562, 0.422, 0.306, 0.367 and 0.380 U cm2, while the total resistances (RT) are 2.181, 1.379, 0.884, 0.865 and 1.180 U cm2, respectively, which are listed in Table 2. Cross-sectional views of SEM micrographs are shown in Fig. 5 with SEM micrographs. From these figures, it seems like all electrolytes are dense enough, which infers that the low OCV of the Cell-2.5 may due to improper sealing of the cell with ceramic paste. To account for the variability of the slurry which affects the electrolyte thickness of the 5 cells (i.e. 23, 18, 16, 13 and 20 mm, respectively), the electrolyte thickness of each cell was adjusted to the same value for comparison. Assuming the conductivity of YSZ is 0.029 S cm1 at 800 C [31], the ohmic resistances of the electrolytes due to different thicknesses would be 0.1586, 0.1241, 0.1103, 0.0896 and 0.1379 U, respectively. If all cells' electrolytes were made to the same thickness of 13 mm, the ohmic resistances of the 5 cells will be 0.4930, 0.3875, 0.2853, 0.3670 and 0.3317 U cm2, respectively, and the trend of change on ohmic resistance is the same with test results. With such adjustment, the electrochemical performances at 800 C were calculated in Table 2, and Cell-2.5 showed the best performance. Since Cell-1.5, Cell-2, Cell-2.5, Cell-3 and Cell-3.5 had same air electrode and same thickness electrolyte after adjustment, it could infer that it was fuel electrodes which affected the electrochemical performance of different cells. The micrographs of fuel electrodes with different amounts of pore former are shown in Fig. 6. In
Table 2 Ohmic resistances (RU), total resistances (RT), peak power densities, electrolyte thickness and their adjustment values for different cells. Cell no.
Cell-1.5
Cell-2
Cell-2.5
Cell-3
Cell-3.5
RU (U cm2) RT (U cm2) Electrolyte thickness (mm) Power density (mW cm2) RU with adjustment (U cm2) RT with adjustment (U cm2) Electrolyte thickness with adjustment (mm) Power density with adjustment (mW cm2)
0.562 2.181 23 317.5 0.4930 2.112 13 327.9
0.422 1.379 18 466.5 0.3875 1.3445 13 478.5
0.306 0.884 16 560.8 0.2853 0.8633 13 574.2
0.367 0.865 13 570.7 0.3670 0.8650 13 570.7
0.380 1.180 20 509.6 0.3317 1.1317 13 531.4
152
J. Zhou et al. / Energy 115 (2016) 149e154
Fig. 6(a), Cell-1.5 shows the lowest porosity which leads to poor power density. With increasing the amount of pore former, the porosity increases and the electrochemical performance is improved. However, after the amount of pore former used reaches the optimal value, three phase boundary decreases and leads to the gradually reduce of the electrochemical performance. Hence, the peak power density of Cell-3.5 is reduced to 509.6 mW cm2 at 800 C. After the above work, the activation process of Cell-2.5 had been studied for its performance stability, which was essentially over one thermal circle and discharged at a constant voltage of 0.7 V at 800 C. Fig. 7 shows the current-voltage (I-V) and current-power density (I-P) curves under different elapsed times. During the discharging at the constant voltage, Cell-2.5 was in activation and stable process. The peak power density of the Cell-2.5 has reached 1263 mW cm2 after 40 h of operation. Fig. 8 illustrates the test result of the Cell-2.5 discharged at 0.7 V over a complete thermal cycle, showing a reasonably good stability. For the button cell was cut off from large area cell, the large cell should has similar
electrochemical performance. The large cells of Cell-2.5 will be used in the future large area stack work.
4. Conclusions Five different amounts of potato starch working as pore former were added into the green fuel electrode with the weight ratio 1.5, 2, 2.5, 3, 3.5%, respectively. Cells with the composition of NiO þ YSZ/ YSZ/LSCF þ CGO were fabricated by aqueous-based tape casting and screen printing. When the amount of starch used is 2.5 wt%, the cell exhibited the best performance. The peak power density registered for this cell was 560.8 mW cm2 at 800 C. After one complete thermal cycle during the activation process of electrode discharging at 0.7 V and 800 C, the peak power density had reached 1263 mW cm2, displaying reasonable good performance stability. The 10 cm by 10 cm Cell-2.5 will be set up to stack in the future work.
Fig. 5. The SEM micrographs of (a) Cell-3.5, (b) Cell-3, (c) Cell-2.5, (d) Cell-2 and (e) Cell-1.5 with different thickness of electrolyte.
J. Zhou et al. / Energy 115 (2016) 149e154
153
Fig. 6. The SEM micrographs of fuel electrodes with different amounts of pore former: (a) Cell-1.5, (b) Cell-2, (c) Cell-2.5, (d) Cell-3 and (e) Cell-3.5.
Fig. 7. The current-voltage (I-V) and current-power density (I-P) curves of Cell-2.5 under different elapsed times.
Fig. 8. The coulometry test of Cell-2.5 at 0.7 V.
154
J. Zhou et al. / Energy 115 (2016) 149e154
Acknowledgments The authors would like to express their gratitude to National Research Foundation and Nanyang Technological University, Singapore for their generous funding support given to SingaporePeking University Research Centre for a Sustainable Low-Carbon Future (SPURc).
[15]
[16] [17] [18]
References [19] [1] Minh NQ. Ceramic fuel-cells. J Am Ceram Soc 1993;76(3):563e88. [2] Appleby AJ. Fuel cell technology: status and future prospects. Energy 1996;21(7e8):521e653. [3] Singhal SC. Science and technology of solid-oxide fuel cells. Mrs Bull 2000;25(3):16e21. [4] Dufour AU. Fuel cells - a new contributor to stationary power. J Power Sources 1998;71(1e2):19e25. [5] Bedringas KW, Ertesvag IS, Byggstoyl S, Magnussen BF. Energy analysis of solid-oxide fuel-cell (SOFC) systems. Energy 1997;22(4):403e12. [6] Tonekabonimoghadam S, Akikur RK, Hussain MA, Hajimolana S, Saidur R, Ping HW, et al. Mathematical modelling and experimental validation of an anode-supported tubular solid oxide fuel cell for heat and power generation. Energy 2015;90:1759e68. [7] Kim KH, Park YM, Kim H. Fabrication and evaluation of the thin NiFe supported solid oxide fuel cell by co-firing method. Energy 2010;35(12): 5385e90. [8] Amedi HR, Bazooyar B, Pishvaie MR. Control of anode supported SOFCs (solid oxide fuel cells): Part I. mathematical modeling and state estimation within one cell. Energy 2015;90:605e21. [9] Bao J, Krishnan GN, Jayaweera P, Perez-Mariano J, Sanjurjo A. Effect of various coal contaminants on the performance of solid oxide fuel cells: Part I. Accelerated testing. J Power Sources 2009;193(2):607e16. [10] Cronin JS, Wilson JR, Barnett SA. Impact of pore microstructure evolution on polarization resistance of Ni-Yttria-stabilized zirconia fuel cell anodes. J Power Sources 2015;196(5):2640e3. [11] Suzuki T, Hasan Z, Funahashi Y, Yamaguchi T, Fujishiro Y, Awano M. Impact of anode microstructure on solid oxide fuel cells. Science 2009;325(5942): 852e5. [12] Yamaguchi T, Shimizu S, Suzuki T, Fujishiro Y, Awano M. Effects of anode microstructure on the performances of cathode-supported micro-SOFCs. Electrochem Solid State Lett 2009;12(11):B151e3. [13] Gamble S. Fabrication-microstructure-performance relationships of reversible solid oxide fuel cell electrodes-review. Mater Sci Technol 2011;27(10): 1485e97. [14] Kennouche D, Chen-Wiegart Y-cK, Cronin JS, Wang J, Barnett SA. Threedimensional microstructural evolution of ni-yttria-stabilized zirconia solid
[20]
[21] [22]
[23]
[24]
[25]
[26]
[27] [28]
[29]
[30]
[31]
oxide fuel cell anodes at elevated temperatures. J Electrochem Soc 2013;160(11):F1293e304. Tsipis EV, Kharton VV. Electrode materials and reaction mechanisms in solid oxide fuel cells: a brief review. J Solid State Electrochem 2008;12(11): 1367e91. Fergus JW. Oxide anode materials for solid oxide fuel cells. Solid State Ion 2006;177(17e18):1529e41. Sun CW, Stimming U. Recent anode advances in solid oxide fuel cells. J Power Sources 2007;171(2):247e60. Chen-Wiegart Y-cK, Kennouche D, Cronin JS, Barnett SA, Wang J. Effect of Ni content on the morphological evolution of Ni-YSZ solid oxide fuel cell electrodes. Appl Phys Lett 2016;108(8):083903. Ramanathan S, Kakade MB. Aqueous slurry processing of monolithic films for SOFC - YSZ, LSM and YSZ-NiO systems. Int J Hydrog Energy 2011;36(22): 14956e62. Fernandez-Gonzalez R, Molina T, Savvin S, Moren R, Makradi A, Nunez P. Fabrication and electrical characterization of several YSZ tapes for SOFC applications. Ceram Int 2014;40(9):14253e9. Moreno V, Hotza D, Greil P, Travitzky N. Dense YSZ laminates obtained by aqueous tape casting and calendering. Adv Eng Mater 2013;15(10):1014e8. Luo LH, Hou BX, Wu YF, Wang CC, Cheng L, Shi JJ, et al. Effect of graphite contents on the electrochemical performance of SOFC prepared by tape casting. Rare Met Mater Eng 2011;40:306e9. Casarin M, Sglavo VM. Influence of processing conditions on the microstructure of NiO-YSZ supporting anode for solid oxide fuel cells. Ceram Int 2015;41(2):2543e57. Choi Y-G, Park J-Y, Song H, Kim H-R, Son J-W, Lee J-H, et al. Microstructurepolarization relations in nickel/gadolinia-doped ceria anode for intermediatetemperature solid oxide fuel cells. Ceram Int 2013;39(4):4713e8. Clemmer RMC, Corbinw SF. Effect of graphite pore-forming agents on the sintering characteristics of Ni/YSZ composites for solid oxide fuel cell applications. Int J Appl Ceram Technol 2011;9(6):1022e34. Xue Y, Guan W, He C, Wang J, Liu W, Sun S, et al. Fabrication of porous anodesupport for planar solid oxide fuel cell using fish oil as a pore former. Int J Hydrog Energy 2016;41(20):8533e41. Drozdz E, Stachura M, Wyrwa J, Rekas M. Effect of the addition of pore former. J Thermal Anal Calorim 2015;122(1):157e66. Horri BA, Selomulya C, Wang H. Electrochemical characteristics and performance of anode-supported SOFCs fabricated using carbon microspheres as a pore-former. Int J Hydrog Energy 2012;37(24):19045e54. Albano MP, Garrido LB, Plucknett K, Genova LA. Influence of starch content and sintering temperature on the microstructure of porous yttria-stabilized zirconia tapes. J Mater Sci 2009;44(10):2581e9. Zhou J, Liu Q, Sun Q, Chan SH. A low cost large-area solid oxide cells fabrication technology based on aqueous Co-tape casting and Co-sintering. Fuel Cells 2014;14(4):667e70. Fergus JW. Electrolytes for solid oxide fuel cells. J Power Sources 2006;162(1): 30e40.
Contents lists available at ScienceDirect
Energy journal homepage: www.elsevier.com/locate/energy
Influence of pore former on electrochemical performance of fuel-electrode supported SOFCs manufactured by aqueous-based tape-casting Juan Zhou b, *, Qinglin Liu b, Lan Zhang b, Zehua Pan a, b, Siew Hwa Chan a, b, ** a b
School of Mechanical and Aerospace Engineering, Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798, Singapore Energy Research Institute at NTU (ERI@N), Nanyang Technological University, 1 CleanTech Loop #06-04, Singapore 637141, Singapore
a r t i c l e i n f o
a b s t r a c t
Article history: Received 15 December 2015 Received in revised form 22 July 2016 Accepted 27 August 2016
The microstructure of a fuel electrode has a significant influence on the whole performance of fuelelectrode supported solid oxide fuel cells (SOFCs) fabricated by aqueous-based tape casting. While in the aqueous-based tape casting process, the fuel electrode porosity, which plays a key role in the final fuel electrode microstructure, mainly comes from the pore former (potato starch in this case). Different contents of starch are added into fuel electrode slurries. When the starch content is 2.5%wt, the cells show the best performance. After one thermal cycle and discharging at a constant voltage of 0.7 V and temperature of 800 C, the peak power density has reached ~1263 mW cm2 with humidified H2 as the fuel and air as the oxidant. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Solid oxide fuel cells (SOFCs) Aqueous-based tape casting Pore former Starch
1. Introduction Solid oxide fuel cells (SOFCs) technology is a cutting-edge technology for directly converting the chemical energy of the fuel into the electrical energy and heat by means of the electrochemical reactions [1,2]. SOFCs technology has many advantages compared with the conventional power plants, such as high efficiency, fuel flexibility, environmentally friendly, low noise level, and etc. [3e5]. SOFCs are mainly divided into four categories, namely the fuel electrode-, air electrode-, electrolyte-, and structural-supported SOFCs [6,7]. Different types of supports have different characteristics, hence exhibiting both advantages and disadvantages. However, the fuel electrode-supported SOFC is considered to be the most mature one compared to the other three [8]. The supporting
List of abbreviation and nomenclature: SOFCs, solid oxide fuel cells; TPB, three phase boundary; OCV, open circuit voltages; I-V, current-voltage; I-P, current-power density; SEM, scanning electron microscope; RU, ohmic resistances; RT, total resistances. * Corresponding author. ** Corresponding author. School of Mechanical and Aerospace Engineering, Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798, Singapore. E-mail addresses: [email protected] (J. Zhou), [email protected] (S.H. Chan). http://dx.doi.org/10.1016/j.energy.2016.08.093 0360-5442/© 2016 Elsevier Ltd. All rights reserved.
fuel electrode provides the mechanical strength to the whole cell, and offers the cell with high electronic conductivity and porosity. The microstructure of the supporting fuel electrode is crucial to the performance of the entire cell [9e12] as gas permeability and reactions strongly depend on microstructural parameters such as porosity, particle size and three phase boundary length [13e15]. Porous nickel-yttria stabilized zirconia (Ni-YSZ) is currently the most common fuel electrode material for SOFCs [16e18]. During the fabrication process, the starting materials of fuel electrode (NiO and YSZ mixture) are prepared by high temperature processing. Then NiO-YSZ is reduced to metallic Ni-YSZ when the fuel electrode is exposed to the fuel environment during the cell operation. In the process of reduction from NiO to Ni, the volume of particles shrunk, which creates some amount of porosity, but this amount of porosity is insufficient to meet the requirements of gas permeability, electrochemical reactions and steam removal. Hence, the pore former is added to increase the porosity. However, if the porosity is lower and higher than the optimal value, the area of the three phase boundary region would not be maximised, which infers that it is crucial to determine the best amount value of porosity with the optimized amount of pore former added into the fuel electrode slurry. Tape casting is a well-known shaping technology for producing large area, thin, uniform and flat ceramic tapes. This technology (organic-based tape casting) is more popular in SOFCs
150
J. Zhou et al. / Energy 115 (2016) 149e154
Table 1 The formula of different fuel electrode slurries with different amounts of poreformer. Chemicals
The ratio of the fuel electrode layer (%)
NiO YSZ Potato starch Poly acrylic acid (PAA) De-ionised water Poly vinyl alcohol, 87e89% partially hydrolysed Poly ethylene glycol (PEG)(Mw 380e420) Glycerol 1-Octanol 2,4,7,9-Tetramethyl-5-decyne-4,7-diol ethoxylate
50 50 1.5, 2, 2.5, 3, 3.5 2 75 10.5 14 14 1.2 0.75
manufacturing. However, organic-based tape casting often involves the use of toxic solvents and hazardous additives, which not only increases the manufacturing cost, but is also harmful to human health and the environment. For this reason, there are increasingly more attentions paid to development of aqueous-based tape casting [19e21]. While in a stable aqueous-based tape casting slurry system, the porosity of fuel electrode is mainly determined by the amount of pore former. There are several papers studied the effect of different pore formers on the properties of the fuel electrode [22e28], while only a handful of literature have been reported with regard to the influence of pore formers on the morphology in the aqueous tape casting [29]. The aim of this work is to study the influence of the pore former (potato starch in this paper) on the microstructure of fuel electrode and its associated electrochemical performance in the aqueous-based tape casting. 2. Experimental 2.1. Fabrication Firstly, the aqueous co-tape casting and co-sintering were used to fabricate fuel-electrode supported half cells. Commercial NiO (J. T. Baker, US) and Zr0.92Y0.08O2-d (YSZ, TOSOH, Japan) were used for preparing the fuel-electrode layer, and the YSZ powder was used to prepare the electrolyte layer. In the aqueous-based tape casting process, potato starch powder (pore former), polyacrylic acid (disperse agent), ammonia (electrostatic dispersant), hydrolysed polyvinyl alcohol (the binder), glycerol and polyethylene glycol (plasticizer), 1-octanol (de-foamer) and 2,4,7,9-tetramethyl-5-decyne-4,7-diol ethoxylate (surfactant) were added. Details of the fabrication process can be found in our
Fig. 2. The picture of (a) 10 cm 10 cm half cell and (b) the button cell.
previous paper [30]. After the NiO þ YSZ/YSZ half-cell was gotten, the La0.6Sr0.4Co0.2Fe0.8O3þd(LSCF)/Ce0.9Gd0.1O2þd (CGO) composite air-electrode with a weight ratio of 1:1 was screen printed on it. To get a stable slurry system, the disperse agent, the electrostatic dispersant, the binder, the plasticizer, the de-foamer and the surfactant need to be in a fixed formula, even though each they can affect the porosity separately, while pore former can be more flexible. Hence, it is much easier to study the influence of the electrode microstructure by changing the amount of pore-former. Table 1 shows different amounts of pore-former used in the fuel electrode slurries with the weight ratio of 1.5, 2, 2.5, 3, 3.5%, respectively. Finally, the 10 cm by 10 cm single cells were fabricated. Fig. 1 shows the cross-section micrographs of the cell with
Fig. 1. The microstructure of single cell with the weight ratio of 2.5%: (a) the low magnification, and (b) high magnification SEM micrograph of the cell's cross-sectional view.
J. Zhou et al. / Energy 115 (2016) 149e154
151
700 C with humidified H2 as the fuel and air as the oxidant. The flow rate of the humidified hydrogen was set at 50 ml min1. The polarisation curve and electrochemical impedance spectrum were obtained using an Autolab PG30/FRA system (Eco Chimie, Netherlands) in the frequency range of 0.1 Hze100 KHz with an excitation potential of 10 mV. A four-probe configuration was adopted in the electrochemical testing. The microstructure of the ceramic samples was characterized by a scanning electron microscope (SEM, JEOL 5600, Japan).
3. Results and discussion
Fig. 3. Typical polarisation curves of voltage and power density versus the current density for different single cells with different amount of pore former at 800 C.
Fig. 4. The impedance spectra of different cells at 800 C.
the weight ratio of 2.5%. 2.2. Measurements To study the influence of different amounts of pore former on electrochemical performance much more effectively, button cells with a diameter of 2.5 cm were cut off from the large area samples. Fig. 2(a) shows the picture of 10 cm 10 cm cell and (b) button cells. The effective area of the air electrode was ~0.5 cm2, and the ceramic paste was used as sealer. The cell performance and electrochemical impedance were measured from 800 C down to
Based on different amounts of pore former in the green fuelelectrode, the cells with 1.5, 2, 2.5, 3, 3.5 wt% potato starches are named as Cell-1.5, Cell-2, Cell-2.5, Cell-3 and Cell-3.5, respectively. Compositions of the cells are all NiO þ YSZ/YSZ/LSCF þ CGO. Fig. 3 shows typical polarisation curves of voltage and power density versus the current density for different single cells with different amounts of pore former at 800 C. The Cell-2.5 and Cell-3 have achieved better electrochemical performance with the peak power density of 560.8 and 570.2 mW cm2 respectively than others. The open circuit voltages (OCV) values are 1.08, 1.1, 1.02, 1.09 and 1.09 V for Cell-1.5, Cell-2, Cell-2.5, Cell-3 and Cell-3.5, respectively. The OCV of Cell-2.5 is slightly lower, while other cells are in good agreement with the theoretical values calculated from the Nernst equation. Fig. 4 shows the impedance spectra of the five cells operated at 800 C. The ohmic resistances (RU) of the five cells are 0.562, 0.422, 0.306, 0.367 and 0.380 U cm2, while the total resistances (RT) are 2.181, 1.379, 0.884, 0.865 and 1.180 U cm2, respectively, which are listed in Table 2. Cross-sectional views of SEM micrographs are shown in Fig. 5 with SEM micrographs. From these figures, it seems like all electrolytes are dense enough, which infers that the low OCV of the Cell-2.5 may due to improper sealing of the cell with ceramic paste. To account for the variability of the slurry which affects the electrolyte thickness of the 5 cells (i.e. 23, 18, 16, 13 and 20 mm, respectively), the electrolyte thickness of each cell was adjusted to the same value for comparison. Assuming the conductivity of YSZ is 0.029 S cm1 at 800 C [31], the ohmic resistances of the electrolytes due to different thicknesses would be 0.1586, 0.1241, 0.1103, 0.0896 and 0.1379 U, respectively. If all cells' electrolytes were made to the same thickness of 13 mm, the ohmic resistances of the 5 cells will be 0.4930, 0.3875, 0.2853, 0.3670 and 0.3317 U cm2, respectively, and the trend of change on ohmic resistance is the same with test results. With such adjustment, the electrochemical performances at 800 C were calculated in Table 2, and Cell-2.5 showed the best performance. Since Cell-1.5, Cell-2, Cell-2.5, Cell-3 and Cell-3.5 had same air electrode and same thickness electrolyte after adjustment, it could infer that it was fuel electrodes which affected the electrochemical performance of different cells. The micrographs of fuel electrodes with different amounts of pore former are shown in Fig. 6. In
Table 2 Ohmic resistances (RU), total resistances (RT), peak power densities, electrolyte thickness and their adjustment values for different cells. Cell no.
Cell-1.5
Cell-2
Cell-2.5
Cell-3
Cell-3.5
RU (U cm2) RT (U cm2) Electrolyte thickness (mm) Power density (mW cm2) RU with adjustment (U cm2) RT with adjustment (U cm2) Electrolyte thickness with adjustment (mm) Power density with adjustment (mW cm2)
0.562 2.181 23 317.5 0.4930 2.112 13 327.9
0.422 1.379 18 466.5 0.3875 1.3445 13 478.5
0.306 0.884 16 560.8 0.2853 0.8633 13 574.2
0.367 0.865 13 570.7 0.3670 0.8650 13 570.7
0.380 1.180 20 509.6 0.3317 1.1317 13 531.4
152
J. Zhou et al. / Energy 115 (2016) 149e154
Fig. 6(a), Cell-1.5 shows the lowest porosity which leads to poor power density. With increasing the amount of pore former, the porosity increases and the electrochemical performance is improved. However, after the amount of pore former used reaches the optimal value, three phase boundary decreases and leads to the gradually reduce of the electrochemical performance. Hence, the peak power density of Cell-3.5 is reduced to 509.6 mW cm2 at 800 C. After the above work, the activation process of Cell-2.5 had been studied for its performance stability, which was essentially over one thermal circle and discharged at a constant voltage of 0.7 V at 800 C. Fig. 7 shows the current-voltage (I-V) and current-power density (I-P) curves under different elapsed times. During the discharging at the constant voltage, Cell-2.5 was in activation and stable process. The peak power density of the Cell-2.5 has reached 1263 mW cm2 after 40 h of operation. Fig. 8 illustrates the test result of the Cell-2.5 discharged at 0.7 V over a complete thermal cycle, showing a reasonably good stability. For the button cell was cut off from large area cell, the large cell should has similar
electrochemical performance. The large cells of Cell-2.5 will be used in the future large area stack work.
4. Conclusions Five different amounts of potato starch working as pore former were added into the green fuel electrode with the weight ratio 1.5, 2, 2.5, 3, 3.5%, respectively. Cells with the composition of NiO þ YSZ/ YSZ/LSCF þ CGO were fabricated by aqueous-based tape casting and screen printing. When the amount of starch used is 2.5 wt%, the cell exhibited the best performance. The peak power density registered for this cell was 560.8 mW cm2 at 800 C. After one complete thermal cycle during the activation process of electrode discharging at 0.7 V and 800 C, the peak power density had reached 1263 mW cm2, displaying reasonable good performance stability. The 10 cm by 10 cm Cell-2.5 will be set up to stack in the future work.
Fig. 5. The SEM micrographs of (a) Cell-3.5, (b) Cell-3, (c) Cell-2.5, (d) Cell-2 and (e) Cell-1.5 with different thickness of electrolyte.
J. Zhou et al. / Energy 115 (2016) 149e154
153
Fig. 6. The SEM micrographs of fuel electrodes with different amounts of pore former: (a) Cell-1.5, (b) Cell-2, (c) Cell-2.5, (d) Cell-3 and (e) Cell-3.5.
Fig. 7. The current-voltage (I-V) and current-power density (I-P) curves of Cell-2.5 under different elapsed times.
Fig. 8. The coulometry test of Cell-2.5 at 0.7 V.
154
J. Zhou et al. / Energy 115 (2016) 149e154
Acknowledgments The authors would like to express their gratitude to National Research Foundation and Nanyang Technological University, Singapore for their generous funding support given to SingaporePeking University Research Centre for a Sustainable Low-Carbon Future (SPURc).
[15]
[16] [17] [18]
References [19] [1] Minh NQ. Ceramic fuel-cells. J Am Ceram Soc 1993;76(3):563e88. [2] Appleby AJ. Fuel cell technology: status and future prospects. Energy 1996;21(7e8):521e653. [3] Singhal SC. Science and technology of solid-oxide fuel cells. Mrs Bull 2000;25(3):16e21. [4] Dufour AU. Fuel cells - a new contributor to stationary power. J Power Sources 1998;71(1e2):19e25. [5] Bedringas KW, Ertesvag IS, Byggstoyl S, Magnussen BF. Energy analysis of solid-oxide fuel-cell (SOFC) systems. Energy 1997;22(4):403e12. [6] Tonekabonimoghadam S, Akikur RK, Hussain MA, Hajimolana S, Saidur R, Ping HW, et al. Mathematical modelling and experimental validation of an anode-supported tubular solid oxide fuel cell for heat and power generation. Energy 2015;90:1759e68. [7] Kim KH, Park YM, Kim H. Fabrication and evaluation of the thin NiFe supported solid oxide fuel cell by co-firing method. Energy 2010;35(12): 5385e90. [8] Amedi HR, Bazooyar B, Pishvaie MR. Control of anode supported SOFCs (solid oxide fuel cells): Part I. mathematical modeling and state estimation within one cell. Energy 2015;90:605e21. [9] Bao J, Krishnan GN, Jayaweera P, Perez-Mariano J, Sanjurjo A. Effect of various coal contaminants on the performance of solid oxide fuel cells: Part I. Accelerated testing. J Power Sources 2009;193(2):607e16. [10] Cronin JS, Wilson JR, Barnett SA. Impact of pore microstructure evolution on polarization resistance of Ni-Yttria-stabilized zirconia fuel cell anodes. J Power Sources 2015;196(5):2640e3. [11] Suzuki T, Hasan Z, Funahashi Y, Yamaguchi T, Fujishiro Y, Awano M. Impact of anode microstructure on solid oxide fuel cells. Science 2009;325(5942): 852e5. [12] Yamaguchi T, Shimizu S, Suzuki T, Fujishiro Y, Awano M. Effects of anode microstructure on the performances of cathode-supported micro-SOFCs. Electrochem Solid State Lett 2009;12(11):B151e3. [13] Gamble S. Fabrication-microstructure-performance relationships of reversible solid oxide fuel cell electrodes-review. Mater Sci Technol 2011;27(10): 1485e97. [14] Kennouche D, Chen-Wiegart Y-cK, Cronin JS, Wang J, Barnett SA. Threedimensional microstructural evolution of ni-yttria-stabilized zirconia solid
[20]
[21] [22]
[23]
[24]
[25]
[26]
[27] [28]
[29]
[30]
[31]
oxide fuel cell anodes at elevated temperatures. J Electrochem Soc 2013;160(11):F1293e304. Tsipis EV, Kharton VV. Electrode materials and reaction mechanisms in solid oxide fuel cells: a brief review. J Solid State Electrochem 2008;12(11): 1367e91. Fergus JW. Oxide anode materials for solid oxide fuel cells. Solid State Ion 2006;177(17e18):1529e41. Sun CW, Stimming U. Recent anode advances in solid oxide fuel cells. J Power Sources 2007;171(2):247e60. Chen-Wiegart Y-cK, Kennouche D, Cronin JS, Barnett SA, Wang J. Effect of Ni content on the morphological evolution of Ni-YSZ solid oxide fuel cell electrodes. Appl Phys Lett 2016;108(8):083903. Ramanathan S, Kakade MB. Aqueous slurry processing of monolithic films for SOFC - YSZ, LSM and YSZ-NiO systems. Int J Hydrog Energy 2011;36(22): 14956e62. Fernandez-Gonzalez R, Molina T, Savvin S, Moren R, Makradi A, Nunez P. Fabrication and electrical characterization of several YSZ tapes for SOFC applications. Ceram Int 2014;40(9):14253e9. Moreno V, Hotza D, Greil P, Travitzky N. Dense YSZ laminates obtained by aqueous tape casting and calendering. Adv Eng Mater 2013;15(10):1014e8. Luo LH, Hou BX, Wu YF, Wang CC, Cheng L, Shi JJ, et al. Effect of graphite contents on the electrochemical performance of SOFC prepared by tape casting. Rare Met Mater Eng 2011;40:306e9. Casarin M, Sglavo VM. Influence of processing conditions on the microstructure of NiO-YSZ supporting anode for solid oxide fuel cells. Ceram Int 2015;41(2):2543e57. Choi Y-G, Park J-Y, Song H, Kim H-R, Son J-W, Lee J-H, et al. Microstructurepolarization relations in nickel/gadolinia-doped ceria anode for intermediatetemperature solid oxide fuel cells. Ceram Int 2013;39(4):4713e8. Clemmer RMC, Corbinw SF. Effect of graphite pore-forming agents on the sintering characteristics of Ni/YSZ composites for solid oxide fuel cell applications. Int J Appl Ceram Technol 2011;9(6):1022e34. Xue Y, Guan W, He C, Wang J, Liu W, Sun S, et al. Fabrication of porous anodesupport for planar solid oxide fuel cell using fish oil as a pore former. Int J Hydrog Energy 2016;41(20):8533e41. Drozdz E, Stachura M, Wyrwa J, Rekas M. Effect of the addition of pore former. J Thermal Anal Calorim 2015;122(1):157e66. Horri BA, Selomulya C, Wang H. Electrochemical characteristics and performance of anode-supported SOFCs fabricated using carbon microspheres as a pore-former. Int J Hydrog Energy 2012;37(24):19045e54. Albano MP, Garrido LB, Plucknett K, Genova LA. Influence of starch content and sintering temperature on the microstructure of porous yttria-stabilized zirconia tapes. J Mater Sci 2009;44(10):2581e9. Zhou J, Liu Q, Sun Q, Chan SH. A low cost large-area solid oxide cells fabrication technology based on aqueous Co-tape casting and Co-sintering. Fuel Cells 2014;14(4):667e70. Fergus JW. Electrolytes for solid oxide fuel cells. J Power Sources 2006;162(1): 30e40.